Discussion on optimizing the configuration method of telecommunication power supply to reduce CAPEX and OPEX

The communication power supply provides DC energy to the communication network, while the battery provides backup energy when the mains fails. On the one hand, power equipment is very important, and on the other hand, its equipment configuration cost also accounts for a certain proportion of the capital construction cost CAPEX. How to reasonably optimize the configuration of the communication power supply is an important part of reducing CAPEX and operation and maintenance costs OPEX.

Thinking about the existing communication power supply configuration method

The current communication power supply configuration method is generally as follows:

Step 1: Calculate the load current I1 (A) = DC load power consumption (W)/48 (V),

Step 2: Calculate the battery capacity C1 (Ah) = 1.25 * KT * T (h) * I1 (A)

Where: KT: Battery time correction coefficient, which is provided by the battery manufacturer and represents the ratio between the actual discharge capacity and the nominal capacity of the battery under different discharge times. Generally obtained by checking the chart.

1.25: Battery capacity compensation coefficient. The actual capacity of the battery will gradually decrease during use. In order to ensure the backup time within the full life cycle of the battery, a compensation coefficient is set.

T: Backup battery delay time. It is set according to the on-site power grid conditions and the importance of the equipment, generally 8 hours.

Step 3: According to the model list of the battery manufacturer, select a battery (whose nominal capacity is C2), and require C2 to be basically equal to or slightly greater than C1.

Step 4: Calculate the battery charging current I2 (A) = KC × C2

Where: KC: battery charging coefficient. According to the requirements of the battery chemical characteristics, it can be between 0.1 and 0.25, that is, the charging time is 4 h ~ 10 h. The specific selection depends on the power grid conditions.

Step 5: Select the specific communication power supply model according to I1 + I2. And require N + 1 backup of the rectifier module.

For example: I1 + I2 is 180A, and a 50A rectifier module is selected, then 4 + 1 rectifier modules are finally configured.

From the above calculation method, we can see that the DC load power consumption has the greatest impact on the configuration of communication power supply and battery. At present, the "maximum power consumption" is used in most cases. And then the question arises, what is the "maximum power consumption"? Is it reasonable to use this maximum power consumption to configure the battery and communication power supply?

The actual power consumption of communication equipment and the concept of average power consumption

For communication equipment such as switches, when a channel is in use, its power consumption is the largest, for example, 10mA, and when it is idle, its power consumption is only about 1mA. In other words, when all channels are in use, the power consumption of this device is the "maximum power consumption". In the actual configuration of the switch, it is impossible to make such an unsecured configuration. Under normal configuration, when the traffic volume is the heaviest, only about 70-80% of the channels are occupied. In other words, the "actual maximum power consumption" is at most 80% of the "maximum power consumption".

Obviously, the "actual maximum power consumption" does not appear 24 hours a day. When the traffic volume is extremely small in the early morning, the power consumption of the switching equipment is only one-tenth of the maximum power consumption.

The same situation also exists in mobile communications.

In this figure, it can be seen that the power consumption of this base station is the highest at 20 o'clock, reaching 4100W (that is, the actual maximum power consumption, and the theoretical "maximum power consumption" of this base station is 4800W). The lowest power consumption is only 1400W. The fundamental reason for such a huge difference is that the base station business volume changes over time, resulting in changes in the actual power consumption of the base station.

Moreover, modern base stations, especially CDMA and 3G base stations, all use power control technology. The number of users, business volume, user distribution and distance, building shielding, base station distribution, etc. will affect the actual power consumption of the base station. These influences make the actual power consumption of the base station further reduce than the theoretical maximum power consumption.

Therefore, if the communication power supply and battery are configured and calculated according to the maximum power consumption of 4800W, more batteries and communication power supplies will be configured, thereby increasing the capital construction cost CAPEX. So

what power consumption should be used to calculate the configuration of communication power supply and battery?

Here we introduce a concept called "average power consumption". That is, the power consumption in a time period is integrated and divided by this time. For example, for the base station shown in Figure 1, its average power consumption within 24 hours a day is about 2700W.

However, in specific project configuration, the concept of "average power consumption during N hours of busy time" should be adopted. Among them, N hours is the desired battery backup time. In general projects, the battery backup is required for 8 hours, so the "average power consumption during 8 hours of busy time" is generally used. In Figure 1, we can see that 14:00 to 22:00 is the period with the highest power consumption, and its "average power consumption during the 8-hour busy period" is approximately 3700 W.

Comparison of the impact of "maximum power consumption" and "average power consumption" on CAPEX and OPEX The impact of

"maximum power consumption" of 4800 W and "average power consumption during the 8-hour busy period" of 3700 W on the configuration of batteries and communication power supplies is shown in Table 1.

Table 1 Comparison of the impact of "maximum power consumption" and "average power consumption during 8 hours of busy hours" on the configuration of batteries and communication power supplies (Note: the battery time correction coefficient KT for 8 hours is 1.08, and the battery charging coefficient KC is 0.15)


Maximum power consumption
Average power consumption during 8 hours of busy hours

Load power consumption value (W)
4800
3700

Load current value I1 (A)
100
77

Battery capacity C1 (Ah)
1080
832

Battery capacity C2 (Ah)
1200 (2 groups of 600Ah)
800 (2 groups of 400Ah)

Charging current I2 (A)
180
120

Total current I1+I2 (A)
280
197

Rectifier module capacity (A)
50
50

Number of rectifier modules (pieces)
7
5



It can be seen that in terms of the configuration of batteries and communication power supplies, the configuration obtained by using the "average power consumption during 8 hours of busy hours" configuration method is more streamlined on the basis of meeting the battery backup time. Roughly reduce equipment investment by 20%.

The reduction in equipment configuration can also use smaller capacity equipment, reduce the size of the equipment and the area it occupies, use a smaller machine room, and reduce infrastructure or rental costs. ZTE has developed several communication power supply equipment based on this idea, such as ZXDU58 S151, which can place the battery pack in the communication power supply rack, reducing the floor space.

Due to the reduction in battery and communication power capacity, the capacity of other related power equipment, such as oil engines, substation/distribution equipment, oil engines, etc. can also be reduced.

The above are all reductions in capital construction costs CAPEX brought about by the use of "average power consumption".

On the one hand, the operation and maintenance cost OPEX is reflected in the fewer spare parts of the rectifier module. On the other hand, it is more reflected in the savings in electricity bills. It is true that the actual power consumption of the communication equipment has not changed in the two algorithms, but the number of configured rectifier modules is different, resulting in the communication power supply working under different load rate conditions in the two cases.

As shown in Table 1, the actual power consumption of the equipment (taking the average power consumption during the 8-hour busy period) is 77A. The capacity of the communication power supply in the two cases is 350A and 250A, respectively, and the load rate is 22% and 31% respectively. When the load rate of modern communication power supply is above 50%, the efficiency is relatively high, and when the load rate is below 50%, the efficiency decreases as the load rate decreases. Here, the difference in load rate of 9% may cause the efficiency of the communication power supply to change by 1-2%, and the difference in efficiency will be directly reflected in the difference of 1-2% in the total electricity bill.

Some matters that need to be noted when applying "average power consumption"

1) From the previous "average power consumption" case, we can see that this average power consumption is abstracted based on the actual existing site measurement data, but a new engineering project cannot obtain these data in advance. At this time, the business volume model of the new project and the historical data model of the power consumption of communication equipment under different business volumes are used to comprehensively estimate an "actual maximum power consumption" data close to the actual data. Then multiply it by 90% as the "8-hour busy hour average power consumption".

2) "8-hour busy hour average power consumption" is only applicable to power estimation in general situations. It is not applicable to the situation where the load power consumption increases due to the large increase in communication business volume during holidays such as the Spring Festival and Mid-Autumn Festival. However, since the power grid is generally more secure during holidays, this relatively makes up for the situation that the battery backup time is reduced due to the increase in load power consumption.

3) Although many bidding projects are required to provide "maximum power consumption", many equipment manufacturers do not provide the maximum power consumption, but the "actual maximum power consumption", so detailed analysis is required in the actual project operation.

4) It should be noted that the configuration obtained by using the "8-hour busy hour average power consumption" algorithm may have a certain impact on battery charging: that is, after a rectifier module fails, the battery charging rate will be slightly lower than the original desired charging rate when the power consumption is the highest. However, this probability and impact are very small and can be ignored in engineering. Take the case in Table 1 as an example. After a module fails, the communication power supply can provide a total current of 200A. However, it happens that the city power supply has just been cut off for several hours and the battery needs to be charged. It happens that it is 20 o'clock when the business is the busiest and the load current is 85A (4100W). Therefore, only 115A of current can be provided to the battery for charging, which is slightly less than the original configuration of 120A, and the charging time is slightly extended. Once the load current drops, the battery charging current will increase immediately.

5) Battery backup time: From Table 1, the configuration battery calculated by the "maximum power consumption" is 1.5 times the "average power consumption during 8 hours of busy time", so its battery backup time is also increased from 8h+ to 12h+, which can indeed increase the system reliability slightly and reduce the probability of business interruption. However, this increase is beyond the design expectation and comes at the cost of increasing CAPEX and OPEX. Its reliability increase is far less than the price paid. On the other hand, the probability of the power outage time of the power grid being between 8 hours and 12 hours is very low.

6) This method of optimizing the configuration of communication power supply and battery by "average power consumption" is also applicable to power systems with energy storage devices such as UPS and solar energy. The effect is particularly obvious for large-capacity and high-cost systems.

Through the communication power supply configuration method and combining the actual communication equipment load characteristics, it can be seen that the actual engineering configuration using the "average power consumption" configuration optimization method can reduce the equipment investment of communication power supply and battery by about 20%, thereby reducing CAPEX and OPEX. This method is also applicable to power equipment such as UPS and solar power supply. r> 77Battery

capacity C1 (Ah)
1080
832Battery

capacity C2 (Ah)
1200 (2 groups of 600Ah)
800 (2 groups of 400Ah)

Charging current I2 (A)
180
120Total

current I1+I2 (A)
280
197Rectifier

module capacity (A)
50
50Number

of rectifier modules (units)
7
5It



can be seen that in terms of the configuration of batteries and communication power supply, the configuration obtained by adopting the "8-hour busy hour average power consumption" configuration method is more streamlined on the basis of meeting the battery backup time. It roughly reduces equipment investment by 20%.

The reduction in equipment configuration can also use smaller capacity equipment, reduce the size and area of ​​equipment, use smaller equipment rooms, and reduce infrastructure or rental costs. ZTE has developed several communication power supply equipment based on this idea, such as ZXDU58 S151, which can place the battery pack in the communication power supply rack to reduce the floor space.

Due to the reduction in battery and communication power capacity, the capacity of other related power equipment, such as oil generators, power transformation/distribution equipment, oil generators, etc. can also be reduced. The above are all reductions in capital construction

costs CAPEX brought about by the use of "average power consumption".