Adaptive neuron PID air conditioning unit performance laboratory temperature control system

Abstract

This invention discloses an adaptive neuron PID air conditioning unit performance test laboratory temperature control system, belonging to the field of temperature control technology. It includes: a temperature acquisition module for real-time acquisition of indoor and outdoor temperature signals in the air conditioning unit performance test laboratory and preprocessing the acquired temperature data; and an adaptive neuron PID control module for comparing the temperature data with a preset target temperature, dynamically adjusting the internal weight coefficients and gain coefficients through an improved self-learning algorithm, and generating a corresponding digital control signal based on the deviation obtained from the comparison. This invention solves the problem of traditional PID controller parameters being difficult to adapt to the large time-delay, time-varying characteristics and strong interference effects in air conditioning test laboratory temperature control by setting a weight coefficient correction method with independent learning rates for each channel and an online gain coefficient adjustment mechanism. This significantly reduces the temperature control fluctuation range, stably meets the temperature control accuracy requirements of test standards, and improves the accuracy of test results.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology, specifically to an adaptive neuron PID air conditioning unit performance test laboratory temperature control system. Background Technology

[0002] The air conditioning unit performance testing laboratory is the core facility for conducting type tests on air conditioning units. When determining the coefficient of performance (COP) of railway vehicle air conditioning units using the air enthalpy difference method, specific temperature and humidity environments must be simulated according to the standard TB / T1804-2017 (e.g., a rated cooling condition requiring an outdoor dry-bulb temperature of 35℃, temperature fluctuations not exceeding ±0.5℃, and stability for more than 1 hour) to accurately measure the core performance indicators of the tested unit, such as cooling capacity and heating capacity. The temperature controlled object in the laboratory is a thermal parameter, possessing inherent characteristics of nonlinearity, large time delay, and slow time variation. It is also susceptible to interference from uncertain factors such as personnel movement, opening and closing of doors and windows, and equipment heat dissipation, placing stringent requirements on the accuracy, stability, and anti-interference capabilities of the control system.

[0003] Traditional PID control algorithms are widely used in industrial control due to their simple structure, strong robustness, and high reliability. However, they have significant limitations in temperature control of air conditioning unit performance testing laboratories: First, PID parameters need to be tuned offline using methods such as the Ziegler-Nichols method, relying on the experience of engineers. Furthermore, parameters tuned in one go cannot adapt to time-varying operating conditions and nonlinear objects, resulting in limited control accuracy. Second, conventional PID uses a linear combination of proportional, integral, and derivative parameters of the deviation to form the control quantity, making it difficult to simultaneously ensure steady-state accuracy and dynamic stability. When faced with large time delays and strong disturbances, problems such as large overshoot, long oscillation periods, and long recovery times easily occur. Third, they are highly dependent on the mathematical model of the controlled object. When model parameters (such as delay time and time constant) change, the control effect drops sharply, resulting in insufficient robustness.

[0004] To address the shortcomings of traditional PID control, existing technologies have proposed intelligent control schemes such as fuzzy PID and neural network PID, but these still have limitations: fuzzy PID's rule base design is complex, making it difficult to accurately match the dynamic characteristics of temperature control; ordinary neural network PID often employs a multi-layered network structure, resulting in high computational load, poor real-time performance, and fixed gain coefficients that cannot be dynamically adjusted based on deviations. This leads to a difficulty in balancing the conflict between adjustment speed and overshoot, failing to meet the high-precision temperature control requirements of air conditioning unit performance testing laboratories. Therefore, an adaptive neural network PID air conditioning unit performance testing laboratory temperature control system is proposed to solve these problems. Summary of the Invention

[0005] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides an adaptive neuron PID air conditioning unit performance test chamber temperature control system, which solves the problems mentioned in the background section.

[0006] Technical solution To achieve the above objectives, the present invention provides the following technical solution: an adaptive neuron PID air conditioning unit performance test chamber temperature control system, comprising: The temperature acquisition module is used to collect indoor and outdoor temperature signals in the air conditioning unit performance test chamber in real time and to preprocess the collected temperature data. The adaptive neuron PID control module compares temperature data with a preset target temperature, dynamically adjusts internal weighting and gain coefficients through an improved self-learning algorithm, and generates a corresponding digital control signal based on the deviation obtained from the comparison. The control output module is used to convert digital control signals into analog control signals that can be recognized by the actuator; The actuator module is used to adjust the cooling capacity or heating power through analog control signals to maintain the temperature stability of the air conditioning unit performance test chamber.

[0007] Preferably, the initialization steps before adjusting the weight coefficients and gain coefficients are as follows: Pre-set the initial values ​​of the three weight coefficients corresponding to the integral action, proportional action, and differential action; Set the initial value of the gain coefficient based on the requirements of normal operating conditions for laboratory temperature control; Set a temperature control sampling cycle to ensure real-time data acquisition and parameter adjustment.

[0008] Preferably, the specific steps for calculating the relevant parameters of the deviation obtained from the comparison are as follows: The current temperature deviation is calculated based on the temperature data and the preset target temperature. The rate of change of temperature deviation is derived based on the current temperature deviation and the temperature deviation of the previous sampling period. By combining the temperature deviations of the current, previous, and previous sampling periods, the second-order change in temperature deviation is obtained.

[0009] Preferably, the specific steps for dynamically adjusting the weighting coefficients are as follows: Based on temperature deviation and the rate of change of temperature deviation, the three weight coefficients are corrected respectively using a preset learning rate; During the correction process, ensure that the adjustment of the weighting coefficients is related to the magnitude and trend of the temperature deviation, so that the integral, proportional, and derivative actions are adapted to the current temperature control requirements. The updated three weight coefficients are normalized to avoid the control effect being affected by a single weight coefficient being too large or too small.

[0010] Preferably, the specific steps for dynamically adjusting the gain coefficient are as follows: Determine whether the sign of the current temperature deviation is the same as that of the temperature deviation in the previous sampling period; If the signs are consistent, it indicates that the temperature change trend has not been reversed. The gain coefficient should be appropriately reduced to avoid overshoot. If the signs are inconsistent, it indicates that the temperature change trend has been reversed. The gain coefficient should be increased appropriately to speed up the response.

[0011] Preferably, the specific steps of the data preprocessing are as follows: Multiple temperature data points are collected continuously according to the set sampling period; Remove the maximum and minimum values ​​from the collected data to avoid extreme data affecting control accuracy; The remaining valid data are averaged, and the resulting average value is used as the valid temperature data for this sampling.

[0012] Preferably, the specific steps for converting the digital control signal into an analog control signal are as follows: Receive generated digital control signals; The digital control signal is converted into a 4-20mA analog current signal using a digital-to-analog converter. The stability of the converted analog signal is checked to ensure that there is no distortion during signal transmission.

[0013] Preferably, the specific steps for adjusting the cooling capacity are as follows: Receive analog control signals; The flow rate of chilled water is changed by adjusting the speed of the chilled water pump through a frequency converter; By adjusting the flow rate of chilled water, the heat exchange capacity of the surface cooler can be precisely controlled, thereby achieving precise regulation of the laboratory's cooling capacity.

[0014] Preferably, the specific steps for adjusting the heating power are as follows: Receive analog control signals; The input power of the electric heater is adjusted by a thyristor voltage regulator; Based on the deviation between the laboratory temperature and the target temperature, the power output of the electric heater is gradually adjusted to avoid temperature fluctuations caused by sudden power changes.

[0015] Preferably, the specific steps for generating the corresponding digital control signal are as follows: Based on the normalized weight coefficients and the corresponding deviation-related parameters, the parameter weighted sum is calculated; Multiply the weighted sum of the parameters by the current gain coefficient to obtain the increment value of the control quantity; The incremental value is added to the control value of the previous sampling period to generate the digital control signal for the current sampling period.

[0016] Beneficial effects The present invention has the following beneficial effects: (1) The adaptive neuron PID air conditioning unit performance test laboratory temperature control system solves the problem that traditional PID controller parameters are difficult to adapt to the large time delay and time-varying characteristics and strong interference in the temperature control of air conditioning test laboratory by setting a weight coefficient correction method with independent learning rate of sub-channels and an online adjustment mechanism of gain coefficient. This greatly reduces the temperature control fluctuation range and can stably meet the requirements of the test standard for temperature control accuracy, thereby improving the accuracy of the test results.

[0017] (2) The adaptive neuron PID air conditioning unit performance test laboratory temperature control system solves the problem of data distortion caused by sensor error and environmental interference caused by setting up a data preprocessing process of multiple temperature data acquisition, extreme value elimination and average calculation. This makes the temperature data input to the control module more reliable, provides accurate basis for subsequent deviation calculation and control signal generation, and ensures the effective operation of the control algorithm.

[0018] (3) The adaptive neuron PID air conditioning unit performance test chamber temperature control system solves the problem of drastic temperature fluctuation caused by sudden changes in cooling capacity and heating power in traditional adjustment methods by setting the execution mode of adjusting chilled water flow by frequency converter and adjusting electric heater power by thyristor voltage regulator. It achieves stable and continuous adjustment of cooling capacity and heating power, maintains the test chamber temperature in a stable state, and reduces the impact of temperature fluctuation on the test process.

[0019] (4) The adaptive neuron PID air conditioning unit performance test temperature control system solves the problem of control imbalance caused by the excessive proportion of a single parameter during the control parameter adjustment process by setting up the linkage operation of weight coefficient normalization processing, weight coefficient correction, and gain adjustment. It adapts the weight of integral, proportional and derivative actions to different temperature control scenarios, enhances the system's adaptability to changes in operating conditions, and enables the system to maintain stable control effect under different test conditions.

[0020] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0021] Figure 1 This is a structural diagram of the adaptive neuron PID air conditioning unit performance test chamber temperature control system of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] This invention provides a technical solution: an adaptive neuron PID air conditioning unit performance test laboratory temperature control system, such as... Figure 1 As shown, it includes: It is used to collect indoor and outdoor temperature signals in the air conditioning unit performance test laboratory in real time. After preprocessing the collected temperature data, it is transmitted to the adaptive neuron PID control module. The specific steps of data preprocessing are as follows: Multiple temperature data points are continuously collected according to the set sampling period, which is set to 0.3s. 15 raw temperature data points are continuously acquired in a single collection process. Remove the two maximum and two minimum values ​​from the collected data to avoid extreme abnormal data affecting control accuracy; The remaining 11 valid data points are averaged, and the resulting average value is used as the valid temperature data for this sampling and transmitted to the adaptive neuron PID control module.

[0024] The specific method for obtaining valid temperature data is as follows: In the formula, This represents the valid temperature data from this sampling, in degrees Celsius (°C). This indicates the number of valid data points used in the average calculation, with a value of 11. Indicates the first The effective temperature raw data, in degrees Celsius (°C), is acquired using a PT100 platinum resistance temperature sensor with a Class B accuracy. The initial error at zero degrees is ±0.35°C, and the measurement error after the temperature rises is ±(0.35+0.002t)°C (t is the measured temperature). The acquired signal is converted into a digital signal by a PCL-818LS A / D converter card with 12-bit conversion accuracy, supporting 16 single-ended or 8 double-ended analog inputs with isolation.

[0025] The adaptive neuron PID control module receives temperature data transmitted from the temperature acquisition module, compares it with the preset target temperature, dynamically adjusts the internal weight coefficients and gain coefficients through an improved self-learning algorithm, and generates the corresponding digital control signal based on the deviation obtained from the comparison. The specific steps for parameter initialization are as follows: Pre-set initial values ​​for three weight coefficients corresponding to integral action, proportional action, and differential action: initial value for integral weight coefficient R=0.0375, initial value for proportional weight coefficient T=2, and initial value for differential weight coefficient Y=24. The weight coefficients are dimensionless and are used to adjust the weights of integral, proportional, and differential actions. The initial value of the gain coefficient, U=0.01, is set based on the normal operating conditions of laboratory temperature control. The gain coefficient has no unit and is used to adjust the response intensity of the control quantity. The temperature control sampling period is set to I=5s to ensure the real-time performance of data acquisition and parameter adjustment. The sampling period unit is seconds (s) to ensure that the controller can respond to temperature changes in a timely manner.

[0026] The specific steps for calculating the deviation-related parameters are as follows: The current temperature deviation is calculated based on the received real-time temperature data and the preset target temperature. The current temperature deviation is obtained as follows: In the formula, This indicates the current temperature deviation, in degrees Celsius (°C). This indicates the preset target temperature, in degrees Celsius (°C). The target temperature range covers various test conditions specified in the TB / T 1804-2017 standard (such as rated cooling condition 35°C, rated heating condition 7°C for heat pumps, etc.). This represents the valid temperature data for the current sampling period, in degrees Celsius (°C). The rate of change of temperature deviation is derived based on the current temperature deviation and the temperature deviation of the previous sampling period. The method for obtaining the rate of change of temperature deviation is as follows: In the formula, This indicates the rate of change of temperature deviation, expressed in degrees Celsius per second (°C / s). This indicates the current temperature deviation, in degrees Celsius (°C). This indicates the temperature deviation from the previous sampling period, in degrees Celsius (°C). This indicates the temperature control sampling period, in seconds (s), with a value of 5s (consistent with the sampling period in parameter initialization).

[0027] By combining the temperature deviations of the current, previous, and previous sampling periods, the second-order change in temperature deviation is obtained.

[0028] The specific method for obtaining the second-order change in temperature deviation is as follows: In the formula, This represents the second-order change in temperature deviation, expressed in degrees Celsius squared per second (°C / s). 2 ); This indicates the rate of change of the current temperature deviation, expressed in degrees Celsius per second (°C / s). This represents the rate of change of temperature deviation in the previous sampling period, in degrees Celsius per second (°C / s). The temperature control sampling period is represented by seconds (s), with a value of 5s. The calculation method of H is consistent with the current deviation change rate (i.e., H=(DG) / I, where G is the temperature deviation of the previous sampling period).

[0029] The specific steps for updating the weight coefficients are as follows: Iterate through each sampling period, obtain the controller output of the previous sampling period, and normalize the output. Based on the current temperature deviation and the rate of change of temperature deviation, the three weight coefficients are corrected respectively using a preset learning rate; During the correction process, the current temperature deviation, the rate of change of temperature deviation, and the second-order change of temperature deviation are normalized. The deviation characteristic parameters with physical units are uniformly converted into dimensionless relative values ​​to ensure dimensional consistency in subsequent weight coefficient correction calculations and avoid logical contradictions caused by direct calculations with parameters of different units. The specific operations are as follows: Current temperature deviation normalization: Take the actual value of the current temperature deviation calculated above and divide it by the preset maximum allowable value of temperature deviation to obtain the normalized current temperature deviation; where the maximum allowable value of temperature deviation is set to 5℃, which is in line with the temperature control accuracy requirement of ±0.5℃ of the "TB / T 1804-2017" standard and reserves sufficient safety redundancy. Temperature deviation change rate normalization: Take the actual value of the temperature deviation change rate calculated above and divide it by the preset maximum allowable value of the temperature deviation change rate to obtain the normalized temperature deviation change rate; where the maximum allowable value of the temperature deviation change rate is set to 2℃ / s, which is determined based on the maximum temperature change rate measured under laboratory conditions. Normalization of the second-order change in temperature deviation: Take the actual value of the second-order change in temperature deviation calculated above, divide it by the preset maximum allowable value of the second-order change in temperature deviation, and obtain the normalized second-order change in temperature deviation; wherein the maximum allowable value of the second-order change in temperature deviation is set to 0.4℃ / s. 2 The maximum second-order rate of change was determined based on laboratory operating conditions and was adapted to the aforementioned parameters.

[0030] All parameters after normalization are dimensionless relative values, which are used for subsequent weight coefficient correction calculations. Ensure that the adjustment of the weighting coefficients is related to the magnitude and trend of the temperature deviation, so that the integral, proportional, and derivative actions are adapted to the current temperature control requirements. The updated three weight coefficients are normalized to avoid the control effect being affected by a single weight coefficient being too large or too small. After normalization, the weight coefficients are fed back into the control algorithm for the calculation of the next control signal.

[0031] The specific method for obtaining the updated normalized weight coefficients is as follows: The integral weight coefficients are obtained as follows: In the formula, This represents the updated integral weight coefficients; This represents the integral weight coefficients before the update; This represents the integral learning rate, with a value of 0.005. This represents the normalized controller output from the previous sampling period. This indicates the normalized current temperature deviation; This represents the normalized rate of change of temperature deviation; The proportional weighting coefficients are obtained as follows: In the formula, This represents the updated proportional weighting coefficients; This represents the proportional weighting coefficient before the update; This represents the proportional learning rate, with a value of 0.01. This represents the normalized controller output from the previous sampling period. This indicates the normalized current temperature deviation; This represents the normalized rate of change of temperature deviation.

[0032] The differential weighting coefficients are obtained as follows: In the formula, Indicates the updated differential weight coefficients; Indicates the differential weighting coefficients before the update; This represents the differential learning rate, with a value of 0.02. This represents the normalized controller output from the previous sampling period. This indicates the normalized current temperature deviation; This represents the normalized rate of change of temperature deviation; The specific steps for normalizing the weight coefficients are as follows: Calculate the absolute value of each of the three updated weight coefficients (integral weight coefficient, proportional weight coefficient, and differential weight coefficient); Add the absolute values ​​of the three weight coefficients together to get the sum of the absolute values ​​of the weight coefficients; The normalized integral weight coefficient, normalized proportional weight coefficient, and normalized differential weight coefficient are obtained by dividing the updated integral weight coefficient, proportional weight coefficient, and differential weight coefficient by the sum of the absolute values ​​of the above weight coefficients. The three normalized weight coefficients are directly used to calculate the weighted sum of subsequent parameters, avoiding control imbalance caused by a single weight coefficient being too large or too small.

[0033] The specific steps for online adjustment of the gain coefficient are as follows: Determine whether the sign of the current temperature deviation is the same as that of the temperature deviation in the previous sampling period; If the signs are consistent, it indicates that the temperature change trend has not been reversed. The gain coefficient should be appropriately reduced to avoid overshoot. If the signs are inconsistent, it indicates that the temperature change trend has been reversed. The gain coefficient should be increased appropriately to speed up the response. The adjusted gain coefficient is used to generate the control signal for the current sampling period, ensuring that the control effect adapts to the temperature change state.

[0034] The specific method for obtaining the adjusted gain coefficient is as follows: In the formula, This represents the adjusted gain coefficient, which is dimensionless. This represents the gain coefficient before adjustment; it has no unit. This indicates the current temperature deviation, in degrees Celsius (°C). This indicates the temperature deviation from the previous sampling period, in degrees Celsius (°C). This function represents the sign of the input value; it outputs 1 when the input value is positive and -1 when the input value is negative. This represents the correction factor, with a value of 0.1 and no unit. This represents the normalized rate of temperature change and has no unit. The normalized parameter for the rate of temperature change is obtained as follows: First, calculate the actual rate of temperature change. Specifically, take the absolute value of the difference between the current temperature deviation and the temperature deviation of the previous sampling period, and then divide it by the temperature control sampling period.

[0035] The current temperature deviation is the difference between the actual collected effective temperature data and the preset target temperature, while the temperature deviation of the previous sampling period is the temperature deviation calculated in the previous collection. The temperature control sampling period is set to 5 seconds. The normalized temperature change rate is obtained by dividing the calculated actual temperature change rate by the maximum allowable temperature change rate.

[0036] The maximum allowable rate of temperature change is set at 2 degrees Celsius per second. This value is consistent with the maximum allowable rate of temperature deviation change to ensure parameter compatibility.

[0037] The specific steps for generating digital control signals are as follows: Based on the normalized weight coefficients and the corresponding deviation-related parameters, the parameter weighted sum is calculated; The weighted sum of parameters is obtained as follows: In the formula, This represents a weighted sum of parameters; , , These represent the normalized integral weight coefficient, proportional weight coefficient, and differential weight coefficient, respectively. This indicates the normalized current temperature deviation; This represents the normalized rate of change of temperature deviation; This represents the normalized second-order change in temperature deviation. The second-order change rate is obtained as follows: First, obtain the current temperature deviation change rate and the temperature deviation change rate of the previous sampling period. The current temperature deviation change rate is obtained by dividing the difference between the current temperature deviation and the temperature deviation of the previous sampling period by the sampling period (5 seconds); the temperature deviation change rate of the previous sampling period is obtained by dividing the difference between the temperature deviation of the previous sampling period and the temperature deviation of the sampling period before that by the sampling period (5 seconds).

[0038] Multiply the weighted sum of the parameters by the current gain coefficient to obtain the increment value of the control quantity; The incremental value of the control quantity is obtained as follows: In the formula, This represents the increment value of the control quantity, in volts (V). This represents the range conversion factor, in volts (V), with a value of 10V (to adapt to the maximum range of 0 to 10V for digital control signals). This represents the adjusted gain coefficient, which is dimensionless. This represents a weighted sum of parameters, without units.

[0039] The increase in this control quantity is superimposed on the control quantity of the previous sampling period to generate the digital control signal for the current sampling period.

[0040] The specific method for obtaining digital control signals is as follows: In the formula, The digital control signal represents the current sampling period; This represents the normalized controller output from the previous sampling period; It represents the incremental value of the control quantity; the value range of the digital control signal is 0 to 10V, which can be adapted to subsequent signal conversion requirements.

[0041] The control output module is used to convert digital control signals into analog control signals that can be recognized by the actuator; The specific steps of signal conversion are as follows: Receive the generated digital control signal, the value range of which is 0 to 10V; The digital control signal is converted into a 4-20mA analog current signal through a digital-to-analog converter. The digital-to-analog converter uses a PCL-726 D / A converter card, which has a 12-bit conversion accuracy and supports multiple analog output ranges of 0-5V, 0-10V and 4-20mA. The stability of the converted analog signal is checked by sampling the analog signal three times consecutively. If the fluctuation value is ≤0.01mA, the signal is determined to be undistorted. If the fluctuation value is greater than 0.01mA, the signal conversion is repeated until the stability requirements are met.

[0042] The specific method for obtaining analog control signals is as follows: In the formula, This represents the converted analog current signal; This represents the received digital control signal; 1.6 is the conversion coefficient in mA / V, and 4 is the reference current value in mA. This formula is derived based on the linear correspondence between digital control signals (0-10V) and analog control signals (4-20mA), ensuring the accuracy of signal conversion.

[0043] The actuator module is used to adjust the cooling capacity or heating power through analog control signals to maintain the temperature stability of the air conditioning unit performance test chamber.

[0044] The specific steps for adjusting the cooling capacity are as follows: Receive analog control signals; The speed of the chilled water pump is adjusted by a frequency converter. The speed of the chilled water pump is linearly related to the analog control signal. The stronger the signal, the higher the pump speed and the greater the chilled water flow. By adjusting the chilled water flow rate, the heat exchange capacity of the surface cooler is adjusted. The surface cooler is the core heat exchange component of the air conditioning system. When the chilled water flow rate increases, the heat exchange capacity increases and the cooling capacity is enhanced, and vice versa, the cooling capacity is weakened, thus achieving precise adjustment of the cooling capacity of the laboratory.

[0045] The specific method for obtaining chilled water flow rate is as follows: In the formula, This indicates the chilled water flow rate, expressed in cubic meters per hour (m³ / h). 3 / h); The received analog control signal is represented in milliamperes (mA); 0.05 is the flow rate adjustment coefficient, measured in cubic meters per hour per milliampere (mA). 3 / (h·mA)); 1 represents the minimum chilled water flow rate, in cubic meters per hour (m³ / h). 3 / h); This formula is based on analog control signals (4-20mA) and chilled water flow rates (1-2m). 3 The linear correspondence between the flow rate and the flow rate ( / h) is derived, and the flow rate adjustment range is 1~2m 3 / h, suitable for the cooling capacity requirements of 80kW (indoor side) and 90kW (outdoor side) in the laboratory.

[0046] The specific steps for adjusting the heating power are as follows: Receive analog control signals; The input power of the electric heater is adjusted by a thyristor voltage regulator. The thyristor voltage regulator supports two triggering modes: phase-shift voltage regulation and variable-cycle zero-crossing power regulation. The output voltage can be dynamically adjusted according to the control signal strength. Based on the deviation between the laboratory temperature and the target temperature, the power output of the electric heater is gradually adjusted to avoid temperature fluctuations caused by sudden power changes. The rated power of the electric heater is 66kW, and it can achieve continuous power adjustment from 0 to 66kW to meet the temperature compensation requirements of the laboratory.

[0047] The specific method for obtaining the input power of the electric heater is as follows: In the formula, This indicates the input power of the electric heater, measured in kilowatts (kW). The received analog control signal is represented in milliamperes (mA); 4.125 is the power regulation coefficient in kilowatts per milliampere (kW / mA); and 16.5 is the power compensation value in kilowatts (kW). The power regulation range of the electric heater corresponding to this formula is 0 to 66 kW, which matches the output characteristics of the thyristor voltage regulator and the rated power of the electric heater.

[0048] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0049] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An adaptive neuron PID air conditioning unit performance test chamber temperature control system, characterized in that, include: The temperature acquisition module is used to collect indoor and outdoor temperature signals in the air conditioning unit performance test chamber in real time and to preprocess the collected temperature data. The adaptive neuron PID control module compares temperature data with a preset target temperature, dynamically adjusts internal weighting and gain coefficients through an improved self-learning algorithm, and generates a corresponding digital control signal based on the deviation obtained from the comparison. The control output module is used to convert digital control signals into analog control signals that can be recognized by the actuator; The actuator module is used to adjust the cooling capacity or heating power through analog control signals to maintain the temperature stability of the air conditioning unit performance test chamber.

2. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific initialization steps before adjusting the weight coefficients and gain coefficients are as follows: Pre-set the initial values ​​of the three weight coefficients corresponding to the integral action, proportional action, and differential action; Set the initial value of the gain coefficient based on the requirements of normal operating conditions for laboratory temperature control; Set a temperature control sampling cycle to ensure real-time data acquisition and parameter adjustment.

3. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps for calculating the relevant parameters of the deviation obtained from the comparison are as follows: The current temperature deviation is calculated based on the temperature data and the preset target temperature. The rate of change of temperature deviation is derived based on the current temperature deviation and the temperature deviation of the previous sampling period. By combining the temperature deviations of the current, previous, and previous sampling periods, the second-order change in temperature deviation is obtained.

4. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps for dynamically adjusting the weighting coefficients are as follows: Based on temperature deviation and the rate of change of temperature deviation, the three weight coefficients are corrected respectively using a preset learning rate; During the correction process, ensure that the adjustment of the weighting coefficients is related to the magnitude and trend of the temperature deviation, so that the integral, proportional, and derivative actions are adapted to the current temperature control requirements. The updated three weight coefficients are normalized to avoid the control effect being affected by a single weight coefficient being too large or too small.

5. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps for dynamically adjusting the gain coefficient are as follows: Determine whether the sign of the current temperature deviation is the same as that of the temperature deviation in the previous sampling period; If the signs are consistent, it indicates that the temperature change trend has not been reversed. The gain coefficient should be appropriately reduced to avoid overshoot. If the signs are inconsistent, it indicates that the temperature change trend has been reversed. The gain coefficient should be increased appropriately to speed up the response.

6. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps of the data preprocessing are as follows: Multiple temperature data points are collected continuously according to the set sampling period; Remove the maximum and minimum values ​​from the collected data to avoid extreme data affecting control accuracy; The remaining valid data are averaged, and the resulting average value is used as the valid temperature data for this sampling.

7. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps for converting digital control signals into analog control signals are as follows: Receive generated digital control signals; The digital control signal is converted into a 4-20mA analog current signal using a digital-to-analog converter. The stability of the converted analog signal is checked to ensure that there is no distortion during signal transmission.

8. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps for adjusting the cooling capacity are as follows: Receive analog control signals; The flow rate of chilled water is changed by adjusting the speed of the chilled water pump through a frequency converter; By adjusting the flow rate of chilled water, the heat exchange capacity of the surface cooler can be precisely controlled, thereby achieving precise regulation of the laboratory's cooling capacity.

9. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps for adjusting the heating power are as follows: Receive analog control signals; The input power of the electric heater is adjusted by a thyristor voltage regulator; Based on the deviation between the laboratory temperature and the target temperature, the power output of the electric heater is gradually adjusted to avoid temperature fluctuations caused by sudden power changes.

10. The adaptive neuron PID air conditioning unit performance test chamber temperature control system according to claim 1, characterized in that, The specific steps for generating the corresponding digital control signal are as follows: Based on the normalized weight coefficients and the corresponding deviation-related parameters, the parameter weighted sum is calculated; Multiply the weighted sum of the parameters by the current gain coefficient to obtain the increment value of the control quantity; The incremental value is added to the control value of the previous sampling period to generate the digital control signal for the current sampling period.

Images