Realizing Sun Tracking Using 80C196KC Single Chip Microcomputer

0 Introduction

Although concentrating solar collectors must track the sun precisely to ensure good results, they still have obvious advantages over ordinary flat solar collectors, especially for medium and high temperature applications.

Various types of devices, from simple to complex, are used for sun tracking, which can be divided into two main categories, namely mechanical systems and electronic control systems. Electronic control systems generally have higher stability and tracking accuracy. Electronic control systems can be further divided into two categories:

1) An analog control system that uses an optical sensor as feedback;
2) A digital system where a computer calculates the sun's position based on a mathematical formula and tracks it using an optical sensor as feedback.

The simulation system based on the sensor's work has poor adaptability and will blindly track the bright spots on the edge of the cloud in cloudy weather, resulting in energy waste and additional wear and tear on the machinery.

Digital systems are generally considered to have higher accuracy and better adaptability, but the systems are complex and expensive.

If the calculation object of the system is simplified appropriately, a low-cost single-chip microcomputer can be used to replace an expensive programmable controller or microcomputer to realize a digital tracking control system. This can greatly reduce the cost of the system while retaining the flexibility and accuracy unique to digital systems.

1 System Description

The tracking strategy of the system described in this paper is to control the movement of the condenser according to the date and time, and use optical sensors to adjust the initial position of the condenser and correct the position of the condenser during operation.

As shown in Figure 1, the system consists of six parts, namely clock, single-chip microcomputer, drive mechanism, encoder, condenser and sensor. The core component of the system is the 80C196KC single-chip microcomputer.

The microcontroller uses the date and time provided by the clock to calculate the expected position of the condenser, compares it with the current position provided by the encoder, and outputs a control signal. The drive device rotates the condenser according to the signal provided by the microcontroller, and at the same time feeds back the running speed or position increment to the microcontroller through the encoder, forming a closed-loop control system.

Since the current position is calculated incrementally, if the calculation of the current position deviates, it cannot be corrected by feedback, resulting in a cumulative position deviation. Therefore, a sensor must be used to monitor whether the position of the condenser deviates from the sun. When it deviates, a correction program is started to eliminate the current position error.

Optical sensors A and B move with the concentrator to provide solar radiation information to the microcontroller. Sensor A uses the pyramid-shaped photovoltaic cell group mentioned in the literature [1]. The four photovoltaic cells are divided into two groups, which provide deviation signals of azimuth and altitude respectively. When the axis of the concentrator points to the sun, it outputs a zero signal; when it deviates, the signal increases with the deviation angle and the increase of the direct solar radiation intensity (Figure 2). Sensor B consists of a photovoltaic cell that receives radiation all day and a photovoltaic cell that blocks direct radiation. It provides a direct solar radiation intensity signal, and the signal increases with the increase of the direct radiation intensity value (Figure 3). All photovoltaic cells selected for sensors A and B should be kept consistent as much as possible.

1) Determine the intensity of direct radiation from the sky and do not start the correction program when the direct radiation is weak, thereby avoiding blind tracking in cloudy weather;
2) Through simple correction operations, reduce or even eliminate the impact of the environment (solar radiation, temperature, etc.) on the deviation signal, so that the corrected deviation signal can be more consistent when the deviation angle is the same.

2 Control System

The 16-bit single-chip microcomputer 80C196KC has strong computing power and can perform basic floating-point operations through programming. Since the position of the sun in the sky can be fully determined by latitude, date and time, accurate tracking can be achieved using this information. The control system block diagram is shown in Figure 4.

As mentioned above, the system uses the sun position calculated by the microcontroller as the expected value input of the condenser position, and the output of the drive device is used as feedback to form a closed-loop system. In order to meet the requirements of stability and accuracy, PD regulator and compensation channel are used respectively.
The transfer function of the drive device can generally be expressed as represents the gain and time constant of the drive device respectively. When only proportional k1 is used for regulation, its closed-loop damping coefficient is the total gain of the closed-loop system, which generally cannot meet the requirements. After adding the PD regulator, ζ＝ can adjust the micro-section differential time constant Td to make the damping coefficient meet the requirements.

Since the change in the sun's running position is basically a ramp input, when only the proportional differential regulator is used, the closed-loop system has a steady-state error ess = 1/K. Increasing the K value of the system can reduce the steady-state error, but a too large K value will make the system's undamped natural frequency close to or greater than the sampling frequency of the microcontroller, causing the system to lose stability. Although the integral term can eliminate the steady-state error of the system, it may cause the system's stability to deteriorate under certain conditions.

Let Z(s) = 1, then the characteristic equation of the original system remains unchanged, and thus the stability does not change. Let the coefficients of the zero-order and first-order terms of the numerator on the right side of the above equation be zero, then the steady-state error of the ramp input is zero.
Let L(s) = L0 + L1 s, and substitute into the numerator on the right side of the above equation, we get:

Since the compensation channel eliminates the static error of the system, the main channel no longer needs to arrange an integral link to eliminate the steady-state error, and becomes a proportional differential control, which is beneficial to the stability of the system. For different controlled objects, different L1, k1 and Td values ​​can be selected to ensure the stability and dynamic characteristics of the system. This parameter adjustment is very simple and reflects the superiority of digital control.

When discretizing the transfer function in the virtual box of Figure 4, the differential term can be replaced by backward difference, and the algorithm is very simple. When a sufficiently small sampling period is used, it can be guaranteed that the discretized system will not lose stability. When the microcontroller uses a 12MHz crystal oscillator, the overflow period of timer 1 is about 87ms, with more than 500,000 state cycles [4], which is sufficient to complete simple calculation tasks, and the position movement of the sun in such a short time can be ignored. Therefore, using the overflow period of timer 1 as the sampling period has the following advantages:

1) It can meet the calculation task. The time analysis of the test system shows that the machine time used for control calculation is less than 15% of the total machine time;
2) When the total system gain K is selected so that the undamped natural frequency of the closed-loop system is not higher than 1Hz, the discretization method can remain stable and achieve sufficient accuracy. The results of the logical analysis of the system show the stability and accuracy of the system.

3 Correction system

Although the above computer control system has high accuracy, the position feedback of the system may still produce errors. And due to the incremental algorithm of the position feedback, this error cannot be detected by the encoder, so it may form a cumulative error. This cumulative error can be corrected by the deviation signal provided by the optical sensor.

Due to the existence of disturbances, the deviation signal will always fluctuate. If correction is performed once the deviation signal is not zero, the system will become sensor-controlled, which has no advantages over analog systems. Setting a dead zone for the corrected deviation signal can greatly reduce the impact of the fluctuation of the deviation signal on the system and increase the stability of the system. The tracking accuracy of the system can be guaranteed by setting the size of the dead zone. When the corrected deviation signal is greater than a given value, that is, it exceeds the dead zone range, and the direct radiation intensity reflected by the reference signal given by sensor B is not lower than a certain threshold, the error correction program is started. The purpose of setting the threshold is to prevent the system from starting the correction process when the direct solar radiation is too weak, that is, when the sun is blocked, so as to avoid blindly tracking the bright spots on the edge of the cloud in cloudy weather. The correction process is implemented in two steps:

1) The optical sensor deviation signal is used to replace the position feedback error E (s) in Figure 4 to form a feedback loop so that the deviation signal tends to 0.
2) When the deviation signal reaches zero, the output position quantity is assigned a value so that the output position quantity is equal to the expected position quantity, and at the same time, the original feedback system is switched back to complete the correction process.

Since the system structure has not changed, the stability of the closed-loop control system formed in the first step above will not change.

4 Conclusion

1) A low-cost digital sun tracking system can be realized by using a single-chip microcomputer.
2) Although simplified, the single-chip microcomputer system has good stability and can achieve fairly good accuracy in sun tracking control, while also having flexibility that analog systems do not have.
3) Using optical sensors, the single-chip microcomputer system can achieve automatic position adjustment.

［1］ William A Lynch，Ziyad M Salameh． Simple electro-optimally controlled dual-axis sun tracker［J］． Solar Energy，1990，45（2）：65－69．
［2］ Soteris A Kalogirou． Design and construction of a one-axis sun-tracking system［J］． Solar Energy，1996，57（6）：465－469．
［3］ Leonard D DJaffe． Test results on parabolic dish concentrators for solar thermal power systems［J］． Solar Energy，1989，42（2）：173－187．
［4］ Sun Hanfang． Intel 16-bit microcontroller［M］． Beijing: Beijing University of Aeronautics and Astronautics Press，1999．