Intelligent temperature adjusting system for industrial water chiller

Abstract

The invention relates to the technical field of non-electrical variable control, and discloses an industrial water chiller temperature intelligent adjusting system, which comprises a refrigeration circulation loop and a control unit, the control unit calculates a steady state control component based on the water outlet temperature deviation, performs differential processing on the refrigerant pressure to extract the gradient, and outputs the gradient. Generating a transient compensation component when the gradient exceeds a dynamic noise threshold; according to the method, the problem of thermal lag is solved by utilizing the pressure wave transmission speed advantage, aggressive response is prevented from triggering low-pressure shutdown through a safety margin constraint mechanism, the pressure safety margin is calculated in real time, a gain coefficient is determined according to a nonlinear model, and a final driving instruction is generated. And the dynamic balance of the control response speed and the system physical security boundary is realized.

Description

Technical Field

[0001] This invention relates to an intelligent temperature control system for industrial chillers, belonging to the field of non-electric variable control technology. Background Technology

[0002] In current precision manufacturing, laser processing, and chemical reaction scenarios, industrial chillers are responsible for providing constant-temperature cooling fluid to non-constant heat loads. They employ an outlet water temperature feedback control mechanism, collecting the outlet water temperature through a temperature sensor and comparing it with a set target value. A PID algorithm is used to calculate the deviation, driving the compressor frequency converter or adjusting the opening of the electronic expansion valve to maintain steady-state water temperature accuracy. Based on temperature deviation closed-loop control logic, basic temperature control requirements are met under conditions of slow load fluctuations or steady-state operation. However, when faced with high-frequency switching of laser generators or transient, severe load disturbances such as pulsed exothermic reactions in reactors, the mainstream temperature feedback method exposes inherent physical limitations. Heat is transferred from the heat source to the cooling water and then transported through the circulation pipeline to the chiller outlet sensor, resulting in an irreducible fluid transport delay of several seconds to tens of seconds. Furthermore, the temperature sensor exhibits an integral effect due to its thermal capacity characteristics. By the time the control system detects a measurable change in the outlet water temperature, the actual disturbance at the load end has already occurred for some time. This spatiotemporal misalignment in the physical transmission mechanism causes the control system response to lag behind the load change, making it difficult to meet the ±0.1℃ accuracy requirements of precision processes.

[0003] Existing technologies attempt to optimize adjustment strategies, but they mostly focus on mechanical structure switching or macroscopic control based on ambient temperature, rather than addressing the transient response of the internal state of the refrigeration cycle. For example, Chinese invention patent CN115540373B discloses an energy-saving industrial chiller that uses artificial intelligence to identify changes in ambient temperature by setting up a partitioned air-cooling component and an internal telescopic component to adjust the volume of the cavity in the evaporator chamber and switch between air-cooling and water-cooling modes, thus solving the problem of avoiding mode misjudgment under different ambient temperatures and achieving energy saving. Such solutions still rely on monitoring of external ambient temperature or system steady-state parameters for control logic. They involve large-scale mode switching and energy efficiency optimization. When faced with millisecond-level severe pulse thermal shocks at the load end, the complex mechanical structure action or ambient temperature feedback adjustment cannot overcome the thermal inertia that causes response lag, and it is difficult to achieve accurate transient compensation of cooling capacity through the extension and contraction of mechanical components.

[0004] Therefore, how to overcome the physical hysteresis of temperature feedback and the dynamic mismatch of refrigerant state changes, and ensure that the compressor can respond quickly to transient loads without triggering low-pressure shutdown, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: An intelligent temperature control system for industrial chillers, comprising:

[0006] The refrigeration cycle loop includes a variable frequency compressor, a condenser, an electronic expansion valve, and an evaporator;

[0007] Temperature acquisition unit, used to acquire the outlet water temperature of industrial chillers;

[0008] The pressure acquisition unit is used to acquire the refrigerant pressure on the suction side of the variable frequency compressor;

[0009] The control unit internally stores preset low-pressure protection thresholds and dynamic noise thresholds, and operates according to the following control rules: Based on the deviation between the outlet water temperature and the preset target temperature, a first control component is calculated to maintain steady-state temperature accuracy; the refrigerant pressure is processed by time-domain differentiation to extract the pressure change gradient value, and the absolute value of the pressure change gradient value is compared with the dynamic noise threshold; when the absolute value of the pressure change gradient value exceeds the dynamic noise threshold, an original second control component is generated to provide transient compensation for the cooling capacity of the refrigeration cycle; the difference between the refrigerant pressure and the low-pressure protection threshold is calculated in real time as the current operating safety margin; the gain coefficient of the original second control component is corrected based on an internally constructed gain correction model. The gain correction model defines a nonlinear mapping rule that shows a positive correlation between the gain coefficient and the safety margin. This nonlinear mapping rule causes the gain coefficient to nonlinearly decrease as the safety margin decreases when it is lower than a preset warning value; the amplitude of the original second control component is weighted and corrected using the gain coefficient to generate the final second control component; based on the first control component and the final second control component, the final drive command for the variable frequency compressor and electronic expansion valve is generated.

[0010] Preferably, when the control unit detects that the pressure change gradient value is positive and exceeds the dynamic noise threshold, it identifies that the system is in a state of sudden load increase; the original second control component includes an instruction to increase the opening of the electronic expansion valve by a first preset step within a preset time, and an instruction to increase the speed of the variable frequency compressor by a second preset step; the control unit also operates a weight decay rule, which, after generating the final drive instruction, gradually reduces the weight of the final second control component in the final drive instruction according to a preset time constant until it returns to zero as the pressure change gradient value falls back.

[0011] Preferably, the dynamic noise threshold is not a fixed value, but is determined in real time by the control unit based on the current operating frequency of the variable frequency compressor; the control unit stores a reference noise spectrum, which limits the base value of pressure fluctuation caused by the inherent vibration of the variable frequency compressor at different operating frequencies; the control unit queries the reference noise spectrum in real time to update the dynamic noise threshold.

[0012] Preferably, during the generation of the final second control component, the control unit continuously monitors the changing trend of the pressure change gradient; when the inflection point feature of the pressure change gradient changing from increasing to decreasing is detected, the control unit identifies this moment as the energy balance point; at the energy balance point, the control unit performs a state locking operation, sets the current cooling capacity determined by the final second control component as the reference cooling capacity for subsequent control, and terminates further dynamic adjustment of the final second control component, instead performing fine-tuning and correction of the reference cooling capacity based on the first control component.

[0013] Preferably, the control unit is also used to calculate the time change rate of the variable frequency compressor's operating frequency in real time; when the second control component drives the variable frequency compressor to accelerate and the time change rate exceeds a preset acceleration threshold, the control unit generates a third control component for the electronic expansion valve based on the time change rate; the third control component is used to drive the opening of the electronic expansion valve to increase at a rate synchronized with the acceleration of the variable frequency compressor, and the increase is positively correlated with the time change rate of the variable frequency compressor's operating frequency, compensating for the decrease in suction pressure caused by the suction action of the variable frequency compressor before the refrigerant evaporation rate in the evaporator increases.

[0014] Preferably, the system further includes a second pressure acquisition unit for acquiring the refrigerant pressure on the condenser side; the control unit internally stores a differential pressure gain correction model, which defines the nonlinear mapping relationship between the feedforward adjustment gain and the system operating differential pressure; the control unit calculates the differential pressure between the refrigerant pressure on the condenser side and the refrigerant pressure on the suction side of the variable frequency compressor in real time, and determines the current differential pressure correction coefficient based on the differential pressure gain correction model; when generating the final second control component, the control unit uses the differential pressure correction coefficient to weight and scale the amplitude of the final second control component.

[0015] Preferably, the control unit is used to monitor the opening adjustment command sent to the electronic expansion valve in real time, and calculate the estimated value of the suction pressure change rate caused by the action of the electronic expansion valve based on the preset valve pressure response model; when extracting the pressure change gradient value, the control unit performs signal decoupling operation, subtracts the estimated value from the measured differential value of the refrigerant pressure, and obtains the net pressure gradient value characterizing the external heat load disturbance; when the absolute value of the net pressure gradient value exceeds the dynamic noise threshold, the control unit generates the original second control component.

[0016] Preferably, the control unit is also used to monitor the suction superheat on the suction side of the variable frequency compressor in real time and calculate the time change rate of suction superheat; during the process of generating the third control component to drive the opening of the electronic expansion valve to increase, the control unit performs logic verification: when the time change rate of suction superheat is detected to be negative and the absolute value exceeds the preset safety threshold, it is identified as a wet compression risk state, and the control unit immediately shields or reverses the third control component, forcibly maintaining or reducing the current opening of the electronic expansion valve until the time change rate of suction superheat returns to the safe range.

[0017] Preferably, the control unit uses a sliding window algorithm to perform time-domain differentiation processing on the refrigerant pressure, and the time length of the sliding window is set to be less than the fluid transport delay time required for the cooling fluid in the refrigeration cycle to be transported from the heat source to the temperature acquisition unit.

[0018] Preferably, the gain coefficients defined in the gain correction model With safety margin The nonlinear mapping rules between them satisfy the following relationship: ,in, To preset warning values, The low-voltage protection threshold is... The preset attenuation index is greater than 1. Used to set the attenuation rate of the gain coefficient when it approaches the low-voltage protection threshold.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] 1. Collect the time-domain variation characteristics of refrigerant pressure on the compressor suction side. Utilize the physical characteristic that the pressure wave transmission speed in the pipeline is faster than the fluid heat transport speed. Identify the load change state before the outlet water temperature changes measurably due to load disturbance and generate compensation commands. Based on the phase advance control mechanism of non-electric variable gradient characteristics, eliminate the control blind zone caused by heat transfer hysteresis and sensor integral effect in the traditional temperature feedback loop. This enables the refrigeration system to establish a preset cooling capacity for transient thermal shock and suppresses dynamic overshoot of water temperature during load change.

[0021] 2. Construct a dynamic coupling relationship between the compressor operating frequency change rate and the throttling device opening adjustment rate. When a sudden increase in load is detected during the compressor acceleration transient process, a synchronous liquid injection command for the throttling device is generated based on the compressor acceleration. The compressor mechanical suction rate and the evaporator refrigerant supply rate are forced to synchronize through the time dimension. This avoids the physical mismatch between the refrigerant phase change response in the evaporator and the compressor mechanical action, and prevents the instantaneous low pressure fault on the suction side or the evaporator from drying out due to the rapid acceleration of the compressor. This ensures the mass flow balance and heat exchange efficiency of the system under the ultra-fast response condition.

[0022] 3. Utilizing the inflection point of the suction pressure change gradient trend as the anchor point for switching between feedforward and feedback control logic, the system accurately locks the physical moment when the cooling output and load demand achieve energy balance by monitoring the zero-crossing characteristic of the second derivative of the pressure. It identifies temperature signals independent of hysteresis, enabling the system to terminate open-loop compensation and smoothly transfer to closed-loop fine-tuning the instant load disturbance is suppressed. This avoids energy excess and secondary temperature oscillations caused by over-compensation or hysteresis exit, achieving a seamless transition from transient burst response to steady-state high-precision maintenance. Furthermore, a decoupling logic based on valve action prediction signals is introduced to calculate the theoretical impact component of the current throttling device action on the suction pressure and separate it from the measured total pressure gradient to obtain the net pressure gradient value characterizing external heat load changes. This endogenous disturbance filtering mechanism decouples the superimposed influence of system adjustment actions and external load disturbances on control variables, preventing erroneous feedforward responses induced by routine throttling device adjustments, and ensuring the logic purity and stability of the multivariable control system when performing adaptive adjustment. Attached Figure Description

[0023] Figure 1 This is a flowchart of the system control logic and signal processing of the present invention;

[0024] Figure 2 This is a graph showing the phase lead characteristic of the pressure signal in this invention.

[0025] Figure 3 This is a schematic diagram of the four functional modules of the intelligent adjustment strategy of this invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. The described embodiments are some embodiments of this invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the protection scope of this invention.

[0027] This invention provides an intelligent temperature control system for an industrial chiller, comprising a refrigeration cycle loop, a temperature acquisition unit, a pressure acquisition unit, and a control unit. The refrigeration cycle loop includes a variable frequency compressor, a condenser, an electronic expansion valve, and an evaporator, forming a complete refrigerant circulation channel. The temperature acquisition unit, such as a high-precision thermistor or platinum resistance thermometer, is located at the chiller outlet to acquire the outlet water temperature in real time. The pressure acquisition unit, such as a pressure sensor, is located in the suction line of the variable frequency compressor to acquire the refrigerant pressure on the compressor suction side in real time. The control unit, such as a microcontroller or digital signal processor, maintains signal connections with the temperature acquisition unit, pressure acquisition unit, and the variable frequency compressor and electronic expansion valve in the refrigeration cycle loop. During steady-state operation, the control unit calculates the first control component for maintaining steady-state temperature accuracy based on the deviation between the outlet water temperature collected by the temperature acquisition unit and the preset target temperature using a proportional-integral-differential algorithm. This first control component is used for fine-tuning the variable frequency compressor and electronic expansion valve to cope with minor heat load fluctuations. Due to the physical lag in temperature feedback control, the control unit runs transient compensation logic, acquiring the original refrigerant pressure sequence from the pressure acquisition unit at a high sampling frequency. For example, at 20Hz, the pressure sequence is processed by time-domain differentiation to extract the pressure change gradient value. The time-domain differential processing is implemented using a sliding window algorithm, with the sliding window duration set to 500 ms. This duration is less than the fluid transport delay required for the cooling fluid to travel from the heat source to the temperature acquisition unit in the refrigeration cycle loop. If it is greater than 5 seconds, the gradient value is physically obtained before the temperature change. To prevent pressure fluctuations caused by compressor vibration from triggering false responses, the control unit stores the absolute value of the pressure change gradient. With dynamic noise threshold Comparison, dynamic noise threshold The value is not fixed and is determined in real time by the control unit based on the current operating frequency of the variable frequency compressor. The control unit stores a reference noise spectrum, which is obtained through offline calibration experiments. This limits the pressure fluctuation baseline value caused by the inherent vibration of the variable frequency compressor at different operating frequencies, such as 30Hz to 90Hz. The control unit queries the reference noise spectrum in real time to update the dynamic noise threshold. Only when If the duration of the state exceeds the confirmation window (e.g., 200ms), the control unit determines that the signal is a thermal load disturbance and generates the original second control component for transient compensation of the cooling capacity of the refrigeration cycle loop.

[0028] In the preferred implementation, to distinguish between pressure changes caused by external load disturbances and system self-regulation, the control unit performs signal decoupling operations when extracting the pressure change gradient value. The control unit monitors and sends the opening adjustment command to the electronic expansion valve in real time, and calculates the estimated rate of change of intake pressure caused by the electronic expansion valve action based on a preset valve pressure response model, such as a simplified linear transfer function. The control unit will convert the measured differential value of the refrigerant pressure. Subtract the estimated value Obtain the net pressure gradient value characterizing the external thermal load disturbance. In this case, the control unit only operates at the net pressure gradient value. The absolute value exceeds the dynamic noise threshold. At that time, the original second control component is generated; after the original second control component is generated, the control unit performs a safety constraint check to prevent over-response under low-voltage conditions from causing shutdown. The control unit internally stores a preset low-voltage protection threshold. For example, 0.2 bar, the control unit calculates the refrigerant pressure in real time. With low voltage protection threshold The difference serves as the current operational safety margin. The control unit determines the corrected gain coefficient of the original second control component based on the internally constructed gain correction model. Gain correction model limits the gain coefficient With safety margin A positively correlated nonlinear mapping rule, when the safety margin Below the preset warning value At times, such as 2.0 bar, the gain coefficient With safety margin The reduction exhibits a non-linear decay. In specific implementations, the non-linear mapping rule satisfies the following relationship: ,in To preset warning values, This represents a value slightly higher than the physical shutdown threshold calculation baseline. A preset attenuation index greater than 1, such as 2, is used to set the attenuation rate of the gain coefficient when it approaches the low-voltage protection threshold. The control unit utilizes the gain coefficient... The amplitude of the original second control component is weighted and corrected to generate the final second control component, which corrects the limitation on the compressor's acceleration rate under low-pressure conditions.

[0029] To improve the system's adaptability to environmental conditions, the system also includes a second pressure acquisition unit for acquiring refrigerant pressure on the condenser side. The control unit internally stores a differential pressure gain correction model, defining a nonlinear mapping relationship between the feedforward adjustment gain and the system operating differential pressure. The control unit calculates the differential pressure between the condenser-side refrigerant pressure and the variable frequency compressor suction-side refrigerant pressure in real time. Based on the differential pressure gain correction model, it determines the current differential pressure correction coefficient. When generating the final second control component, the control unit uses the differential pressure correction coefficient to reweight and scale the amplitude of the final second control component or the original second control component again to compensate for the drift in refrigerant pressure response characteristics caused by changes in environmental conditions. When the control unit detects a positive pressure change gradient value... When the dynamic noise threshold is exceeded, the identification system is in a state of sudden load increase. At this time, the original second control component includes the following instructions: increase the opening of the electronic expansion valve by a first preset step size within a preset time, such as increasing the opening by 10%; and increase the speed of the variable frequency compressor by a second preset step size, such as an instantaneous increase of 20Hz. The instructions are used to pre-increase the refrigerant flow and enhance the cooling capacity before the water temperature rises, thus neutralizing the impending thermal shock. When the system is in a state of sudden load increase, during the acceleration process of the variable frequency compressor driven by the second control component, there is a dynamic mismatch problem where the compressor suction rate is faster than the evaporator evaporation rate. To solve this problem, the control unit calculates the rate of change of the variable frequency compressor's operating frequency over time in real time. When the rate of change over time When the preset acceleration threshold is exceeded, the control unit determines the acceleration rate based on the time change rate. Generate the third control component for the electronic expansion valve The third control component This is used to drive the electronic expansion valve opening to increase synchronously with the acceleration rate of the variable frequency compressor, and the increase magnitude is related to the rate of change of the variable frequency compressor's operating frequency over time. There is a positive correlation. Before the refrigerant evaporation rate in the evaporator increases, the active compensation for the decrease in suction pressure caused by the suction action of the variable frequency compressor is used. To prevent the risk of wet compression caused by synchronous liquid injection, the control unit monitors the suction superheat of the variable frequency compressor in real time. Calculate the rate of change of intake superheat over time. Generate the third control component During the process of increasing the opening of the electronic expansion valve, the control unit performs a logic check: when the rate of change of intake superheat over time is detected... Negative and the absolute value exceeds the preset safety threshold At times, such as If the system identifies a wet compression risk condition, the control unit immediately disables or reverses the third control component. The current opening of the electronic expansion valve is forcibly maintained or reduced until the rate of change of intake superheat time returns to a safe range.

[0030] The transient compensation exit mechanism is used to achieve smooth control switching. In the preferred implementation, the control unit continuously monitors the pressure change gradient trend during the generation of the final second control component. When the pressure change gradient is detected... or When the inflection point characteristic changes from increasing to decreasing, this inflection point characteristic corresponds to the zero-crossing point of the second derivative. The control unit identifies this moment as the energy balance point, indicating that the rate of increase in cooling capacity and the thermal shock rate have reached a balance. The control unit performs a state lock operation at the energy balance point, and the current cooling capacity is determined by the final second control component and set as the reference cooling capacity for subsequent control. This cooling capacity is determined by the current compressor frequency and valve opening. The final second control component is dynamically adjusted, and the reference cooling capacity is fine-tuned based on the first control component. In another implementation, the control unit runs a weight decay rule. After generating the final drive command, as the pressure change gradient value falls back, it gradually reduces the weight of the final second control component in the final drive command according to a preset time constant, such as 3 seconds until the weight returns to zero, and smoothly restores the weight of the first control component. Finally, based on the first control component and the final second control component, in the decoupling operation of the pressure change gradient extraction signal, the control unit adopts a recursive correction strategy to maintain temporal causality and calculates the current moment. Estimated rate of change of inspiratory pressure Call the previous sampling period The electronic expansion valve opening command is issued, and the theoretical pressure fluctuation value caused by the valve action at the current moment is calculated based on the preset valve pressure response model; the value obtained from the hysteresis calculation is then... Based on the current measured total pressure gradient Subtracting this yields the net pressure gradient characterizing the abrupt change in external thermal load. To eliminate the real-time coupling of algebraic loops between closed-loop control variables; when identifying the inflection point of the pressure change gradient from increasing to decreasing to determine the energy balance point, the control unit executes trend confirmation logic to filter out high-frequency noise interference from the sensor, and the judgment condition is set as the continuous first-order difference value of the pressure change gradient. The sampling period remains negative and the cumulative decrease exceeds the preset noise tolerance. Example Choose 3, noise margin Set as The calibration is based on the standard deviation of the measured pressure gradient signal during steady-state operation of the system. When the energy balance point is three times the value of the threshold above, the energy balance point is confirmed to have reached the state of execution and locking operation is performed; for dynamic noise thresholds... and noise tolerance On-site calibration: The system runs an initialization self-test program, the compressor maintains its minimum speed and the electronic expansion valve opening is fixed at zero load, and 60 seconds of continuous suction pressure data are collected to calculate the gradient statistical distribution; the distribution... quantiles are... The boundary is used as the reference noise base, and the dynamic noise threshold is written into it. Initial lookup table and noise tolerance The register, serving as the quantization boundary between real load disturbances and inherent mechanical vibrations, generates the final drive commands for adjusting the operating frequency of the variable frequency compressor and the opening of the electronic expansion valve through weighted summation or logical arbitration.

[0031] Example 1: This example illustrates the application of an intelligent temperature control system for an industrial chiller in a specific industrial scenario. In precision laser processing, the industrial chiller system provides a constant-temperature cooling fluid to a high-frequency on / off laser generator, requiring the cooling fluid temperature to remain constant at 20.0°C. The fluctuation range does not exceed ±0.1 At time t0, the system is in steady-state operation, the laser generator is in standby mode, the heat load is extremely low, and the control unit, based on the first control component, maintains the variable frequency compressor at a low frequency of 35Hz, with the outlet water temperature stable at 20.0℃. At this time, the refrigerant pressure on the suction side The pressure stabilizes at 0.5 bar, which corresponds to a safety margin. The reading is 0.3 bar, which is lower than the preset warning value. At time t1, the laser generator instantaneously starts and enters full-power pulse operation mode. Thermal shock is applied to the evaporator, causing the refrigerant inside the evaporator to boil. The pressure acquisition unit monitors the refrigerant pressure on the suction side. As the pressure begins to rise rapidly, the control unit extracts the pressure change gradient value through time-domain differentiation. The absolute value exceeds the dynamic noise threshold determined based on the current 35Hz frequency at time t1+200ms. The system detected a sudden increase in load. Due to fluid transport delays and the thermal capacity of the temperature sensor, the outlet water temperature reading from the temperature acquisition unit remained at 20.0. The control unit generates the initial second control component, which includes commands to drive the compressor to accelerate and the electronic expansion valve to open. Before generating the final second control component, the control unit performs a safety margin check, taking into account the current safety margin. Only 0.3 bar, below the warning value The control unit is based on the internal gain correction model. Calculate the low gain coefficient The control unit utilizes the gain coefficient The compressor acceleration command amplitude in the original second control component is weighted and corrected. The final second control component is modified as follows: the opening of the electronic expansion valve is increased to introduce refrigerant, and the speed of the inverter compressor is slightly increased to avoid triggering a low-pressure shutdown due to excessively low instantaneous suction during low-pressure start-up. As the electronic expansion valve opens and refrigerant floods in, the refrigerant pressure on the suction side increases. Rebounded to a safe range, with a safety margin It then increases and exceeds the warning value. Gain coefficient The value is restored to 1. At this point, the control unit releases the compressor acceleration limit and, including the full-amplitude command, the final second control component drives the inverter compressor to rapidly increase its frequency to cope with continuous thermal load shocks. The control unit calculates the rate of change of the inverter compressor's operating frequency over time. At that time, the value was detected to exceed the preset acceleration threshold.

[0032] To prevent the risk of low pressure caused by the mismatch between the high-speed suction of the variable frequency compressor and the evaporation rate of the evaporator, the control unit is based on the time change rate. Generate the third control component This component drives the electronic expansion valve opening to increase synchronously with the compressor acceleration rate. At time t1+2s, the control unit performs logic verification and monitors the rate of change of suction superheat over time. A negative value occurs when the absolute value exceeds the preset safety threshold. The control unit determines that there is a risk of wet compression and immediately temporarily disables the third control component. The system forcibly maintains the current opening of the electronic expansion valve until t1+2.5s, when the rate of change of suction superheat over time returns to a safe range. The third control component then resumes execution to ensure mass flow balance and prevent compressor liquid slugging. At t1+4s, the control unit continuously monitors the pressure gradient. Trend detected Upon noticing the inflection point where the cooling rate transitions from increasing to decreasing, the control unit identifies this moment as an energy balance point and immediately executes a state lock operation. It sets the current cooling capacity as the reference cooling capacity for subsequent control and terminates the final dynamic adjustment of the second control component. Control is smoothly transferred back to the first control component. Throughout the entire dynamic response process from t0 to t4s, the maximum fluctuation in the outlet water temperature from the temperature acquisition unit is 20.08°C. It remained within ±0.1 Within the required accuracy, the system suppresses transient thermal shock without the risk of low-pressure shutdown or wet compression.

[0033] Example 2: This example verifies the actual control effect, system safety, and stability of the intelligent temperature regulation system for industrial chillers according to the present invention in response to transient thermal load shocks. A standardized experimental verification platform was built to compare the performance indicators of experimental groups using different control strategies under the same operating conditions. The experimental platform was built in a laboratory with controllable enthalpy difference in ambient temperature. The test object was a 5kW variable frequency industrial chiller equipped with a variable frequency compressor, electronic expansion valve, plate heat exchanger, and air-cooled condenser. The temperature acquisition unit used was a PT100 platinum resistance thermometer, installed at the evaporator outlet, with a sampling frequency set to 1Hz and a measurement accuracy of ±0.05. The pressure acquisition unit uses a piezoresistive pressure sensor, installed at the compressor intake port, with a sampling frequency set to 20Hz and a measurement accuracy of ±0.5%FS. The control unit uses an ARM Cortex-M4 core-based microcontroller, running the control logic of this invention. The load simulation device is a programmable electric heater, generating a controlled step thermal load. Three comparative control strategies were set up, corresponding to three experimental groups: Control group 1 used traditional PID control: based on the temperature deviation of the outlet water collected by the temperature acquisition unit, the first control component was calculated to drive the compressor and electronic expansion valve, without enabling pressure gradient feedforward logic; Control group 2 used unconstrained feedforward control: based on control group 1, it enabled control based on the pressure change gradient value. Feedforward control generates the original second control component, but the strategy does not incorporate a safety margin-based approach. The gain correction model does not incorporate the rate of change of intake superheat over time. The wet compression blocking mechanism of this invention is a complete control scheme: enabling features including pressure gradient extraction and based on safety margin. Nonlinear gain correction, synchronous compensation based on compressor acceleration throttle valve and wet compression blocking mechanism, and complete control logic based on second derivative zero-crossing state locking and smooth switching; Experimental conditions: constant ambient temperature 35°C Initially, the load simulation device outputs 1kW, the system operates in steady state, and the outlet water temperature remains stable at 20.0℃. At time T=0, the output power of the load simulation device increases to 5kW, the simulated laser generator starts instantaneously, and the data is recorded after running continuously for 10 minutes. Table 1 shows the comparison data of key performance indicators of the three experimental groups under the above conditions.

[0034] Table 1: Comparison of Dynamic Response Performance and Safety Indicators

[0035]

[0036] The technical analysis of the data in Table 1 is as follows: Control group 1 relies on a temperature signal with thermal hysteresis. When the temperature rises, thermal shock causes the water temperature to deviate significantly from the target value, with the maximum temperature deviation reaching +2.5. The recovery time is as long as 180 seconds, which cannot meet the ±0.1 standard of precision manufacturing. Temperature control requirements confirmed that feedback control alone could not solve the physical spatiotemporal misalignment problem. Control group 2 introduced pressure gradient feedforward, preemptively driving the compressor to accelerate rapidly and the electronic expansion valve to open significantly before temperature changes. Although theoretically the temperature control response was extremely fast, in the experiment, the compressor's rapid suction caused the suction pressure to drop to 0.1 bar in a short time, below the preset low-pressure protection threshold of 0.2 bar, triggering a low-pressure alarm and shutdown. Simultaneously, the unrestrained opening of the electronic expansion valve caused a large influx of liquid refrigerant, instantly reducing the suction superheat to 0.0. C.

[0037] The invention group detected pressure change gradient values. Exceeding the dynamic noise threshold Subsequently, when the feedforward response is initiated and the final second control component is generated, the control unit calculates the safety margin in real time. Using a gain correction model to nonlinearly attenuate the compressor acceleration command, data shows that the lowest suction pressure in this invention group is 1.8 bar, which is within the safe range. When the control unit monitors the rate of change of suction superheat over time... When exhibiting a sharp negative change, the logic verification mechanism limits the driving amplitude of the third control component on the electronic expansion valve, maintaining the minimum suction superheat at 3.5. To avoid wet compression, the present invention group ultimately suppressed the maximum temperature deviation to +0.15°C. Within 35 seconds, the system quickly returned to steady state without any system failure. Experimental results show that this invention utilizes the physical characteristic that the refrigerant pressure wave propagation speed is faster than the fluid heat transport speed, combined with a dynamic constraint mechanism based on safety margin and superheat trend, to simultaneously achieve advanced response to transient thermal load and strict adherence to the system's physical safety boundary within a single control architecture, thus resolving the inherent contradiction between response speed and operational stability in traditional control schemes.

[0038] Example 3: This example combines Figures 1 to 3 A description of an intelligent temperature control system for an industrial chiller, such as... Figure 1 As shown, the pressure acquisition unit collects the inhalation pressure in real time. After time-domain differentiation and signal decoupling processing by the pressure gradient extraction module, the pressure gradient is generated. ,Should The signal is used, on the one hand, to determine the dynamic noise threshold compared with the reference noise spectrum. The system performs a discrimination process, triggering the generation of a second control component when the threshold is exceeded. This component is also used by the safety margin gain correction module, which calculates the gain coefficient K based on the safety margin M to correct the second control component. Simultaneously, the outlet water temperature collected by the temperature acquisition unit is sent to the steady-state control logic module, which calculates the first control component based on the temperature deviation. Furthermore, the system monitors the frequency change rate. The third control component is generated by the synchronous liquid injection module. Finally, the first control component, the modified second control component, and the third control component are sent to the drive command generation and arbitration unit. This unit integrates the components to generate a frequency command to drive the variable frequency compressor to perform frequency regulation and generates an opening command to drive the electronic expansion valve to perform opening regulation.

[0039] like Figure 2 As shown in the figure, the horizontal axis represents time in seconds, the left vertical axis represents the pressure gradient in bars / s, and the right vertical axis represents the temperature deviation. The figure clearly shows that this represents the pressure gradient. It responds quickly and rises after a load disturbance, crossing the dynamic noise threshold between 1.0 and 1.5 seconds. It reaches its peak at 3.0 seconds, representing the deviation in outlet water temperature. Due to the thermal hysteresis effect, the response is significantly delayed, reaching its peak at 3.5 seconds. This timing difference confirms the pressure gradient. Relative to temperature deviation It has phase lead characteristics, providing valuable early warning time for the control system; such as Figure 3 As shown, the system is designed around the central goal of intelligent temperature regulation of industrial chillers. It includes a sensing and feedforward module, which includes functions such as time-domain differential of suction pressure, dynamic noise threshold determination, and decoupling of valve action signals; a safety boundary bundle module, which includes functions such as low-pressure protection threshold, safety margin calculation, and nonlinear gain correction; a transient response coordination module, which includes functions such as synchronous liquid injection of pressure valves, frequency change rate monitoring, and wet compression blocking mechanism; and a steady-state smooth switching module, which includes functions such as energy balance point locking, second derivative zero-crossing recognition, and control weight attenuation.

[0040] Example 4: This example provides a dynamic noise threshold for calibrating pressure gradient feature extraction. Standardized engineering procedures eliminate uncertainties in parameter setting. These procedures apply to the initial configuration or periodic calibration phases of the intelligent temperature control system for industrial chillers of this invention. The initial input definition for the calibration process is: an industrial chiller equipped with the control unit and pressure acquisition unit of this invention, with the chiller in a state of no external heat load connection or connected to a constant zero-load simulator, and the ambient temperature maintained at [temperature value missing]. Under standard operating conditions, the sampling frequency of the pressure acquisition unit is set to 20Hz, and the signal is filtered by a low-pass filter to remove high-frequency electromagnetic interference above 50Hz. The core of the calibration process lies in constructing a reference noise spectrum, which establishes a quantitative mapping relationship between the operating frequency of the variable frequency compressor and the baseline value of the suction pressure fluctuation. The specific execution steps are as follows: Start the variable frequency compressor and set the operating frequency to the lowest operating frequency. For example, at 30Hz, after the system has stabilized, such as after running for 5 minutes, the control unit will begin collecting data for a duration of [duration missing]. For example, 60-second inspiratory pressure time series data The control unit collects data. Performing sliding window differentiation operations that are completely consistent with real-time control logic, the pressure change gradient sequence at this frequency is calculated. Control unit calculation The statistical characteristic value of a sequence is specifically calculated by determining the maximum absolute value or the 99.7th percentile of the sequence. In principle, this statistical characteristic value is defined as the intrinsic noise floor value at the current frequency. To ensure control robustness, this base value is multiplied by a safety factor. ,like The dynamic noise threshold at the current frequency is obtained. The control unit gradually increases the compressor's operating frequency in preset frequency steps, such as 5Hz, until the maximum operating frequency is reached. For example, at 90Hz, for each discrete frequency point, steps one through three are repeated to obtain a series of corresponding dynamic noise threshold data points. The control unit uses linear interpolation or polynomial fitting methods to fit the discrete data points into a continuous frequency threshold curve. This curve data table is stored in non-volatile memory as a reference noise spectrum. During real-time operation, the control unit consults this spectrum based on the current measured compressor operating frequency to determine the current dynamic noise threshold. .

[0041] Example 5: This example provides standardized engineering calibration procedures for constructing a differential pressure gain correction model and a valve pressure response model. These models ensure the control accuracy and stability of the industrial chiller temperature intelligent regulation system under different environmental conditions and different regulation actions. The construction procedure for the differential pressure gain correction model is as follows: Place the industrial chiller system in a test chamber with controllable ambient temperature and connect it to a simulated load source capable of outputting a stable heat load; set the ambient temperature to a low-temperature operating value, such as 10°C. After the system stabilizes at the target water temperature, record the system operating pressure difference at this time. In this state, a preset step thermal shock is applied to the simulated load source, such as an instantaneous increase of 1kW, and the peak value of the gradient of the intake pressure change measured by the pressure acquisition unit is recorded. Set the ambient temperature to a high-temperature operating value, such as 45°C. Repeat the above steps and record the corresponding high voltage difference. Peak pressure gradient caused by the same step thermal shock under this operating condition By comparison and The difference is used to establish a nonlinear mapping relationship between the pressure gradient response sensitivity and the system operating pressure difference, which is then solidified into a pressure difference gain correction model and stored in the control unit. This model is used to weight and scale the amplitude of the final second control component during real-time operation.

[0042] The procedure for constructing the valve pressure response model is as follows: The industrial chiller system is operated under a constant heat load, such as 2kW and a constant variable frequency compressor frequency of 50Hz, until the outlet water temperature and suction pressure reach a stable state. Under this stable condition, the control unit temporarily suspends all closed-loop control logic. At time t0, it actively sends a preset opening adjustment command to the electronic expansion valve, such as increasing the opening by 5 steps. Within a preset time window after time t0, such as within 2 seconds, the control unit collects the suction pressure sequence at high frequency and calculates the pressure change gradient value of this sequence. Since this pressure change is triggered by the action of the electronic expansion valve, this gradient value is calibrated as the estimated suction pressure change rate under the corresponding operating condition. By repeating this injection test at different steady-state operating points, one or more structures describing valve action and... The relationship lookup table or function model is stored in the control unit and used to perform signal decoupling operations during real-time operation.

[0043] Example 6: This example provides a standardized engineering procedure for on-site adaptive tuning and optimization of core control parameters of the control unit under different installation environments and load characteristics. By performing controlled excitation tests after the initial system deployment or replacement of major components, the real-time response characteristics of the system are obtained, and the attenuation index in the gain correction model is calibrated. and the safety threshold in wet compression blocking logic Initial input definition for the tuning process: An industrial chiller system that has been physically installed, with piping connected and refrigerant charge meeting the nominal value, connected to the actual terminal heat load equipment. Before starting the tuning, ensure the ambient temperature is within the specified range. to Within the effective operating range, the cooling water circulation system is activated and operates stably. The first stage of the tuning process involves identifying the system's open-loop response characteristics. The control unit temporarily freezes the PID closed-loop regulation, sets the variable frequency compressor to a constant reference frequency, such as 40Hz, and locks the electronic expansion valve at a constant opening until the suction pressure... After the outlet water temperature stabilizes, the control unit sends an amplitude of [value missing] to the electronic expansion valve. Step opening command, continuously record inhalation pressure The step response curve is used to extract the process gain of the refrigeration system under the current operating condition by fitting the response curve. and time constant .

[0044] The second stage of the tuning process is the attenuation index. Adaptive calibration, based on process gain identified in the first stage. The control unit calculates the initial attenuation index. The control unit performs a set of closed-loop load disturbance simulation tests: while maintaining a constant target water temperature, a series of virtual load signals with increasing amplitude are artificially introduced, or equivalent pressure disturbances are generated by controlling the rapid action of the electronic expansion valve. During each disturbance, the system records the intake pressure. Downward range and recovery time ,like Exceeding the preset safety margin limit indicates an overreaction, and the control unit increases its capacity. Value; if Exceeding the preset response time limit indicates lag, and the control unit reduces... The value is determined through iterative optimization. and Set value within the optimal equilibrium range The third stage of the tuning process is the wet compression safety threshold. On-site calibration was performed when the system was in steady-state operation and the intake superheat was... Within the normal range, such as to The control unit performs a rapid liquid detection operation: it drives the electronic expansion valve to open at a preset high rate until the intake superheat is detected. When a clear inflection point appears or the exhaust temperature shows a sharp drop, record the rate of change of intake superheat over time at that moment. The peak value is multiplied by a safety factor, such as 0.8, and set as the specific wet compression safety threshold for the current system. The operation ensures that the threshold setting matches the current pipe length, heat exchanger efficiency, and actual refrigerant charge; finally, the parameters obtained through on-site tuning will be... and Write the non-volatile configuration area of ​​the control unit as the operating baseline for the device in the current application environment.

[0045] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An industrial water chiller temperature intelligent regulating system, characterized in that, The application relates to a control method for a refrigeration cycle circuit of an industrial water chiller. The refrigeration cycle circuit comprises a variable frequency compressor, a condenser, an electronic expansion valve and an evaporator. A temperature acquisition unit is arranged to acquire the outlet water temperature of the industrial water chiller. A pressure acquisition unit is arranged to acquire the refrigerant pressure at the suction side of the variable frequency compressor. A control unit is internally stored with a preset low-pressure protection threshold value and a dynamic noise threshold value, and is provided with the following control rules: based on the deviation between the outlet water temperature and a preset target temperature, a first control component for maintaining the steady-state temperature accuracy is calculated; the refrigerant pressure is subjected to time-domain differential processing to extract a pressure change gradient value, and the absolute value of the pressure change gradient value is compared with the dynamic noise threshold value; when the absolute value of the pressure change gradient value exceeds the dynamic noise threshold value, an original second control component for transiently compensating the refrigerating capacity of the refrigeration cycle circuit is generated; the difference between the refrigerant pressure and the low-pressure protection threshold value is calculated in real time as a safety margin of the current operation; A gain correction model internally constructed by the control unit is used to correct the gain coefficient of the original second control component, the gain correction model defines a nonlinear mapping rule between the gain coefficient and the safety margin, and the nonlinear mapping rule causes the gain coefficient to be nonlinearly attenuated with the decrease of the safety margin when the safety margin is lower than a preset warning value; The gain coefficient is used to weight and correct the amplitude of the original second control component to generate a final second control component, and based on the first control component and the final second control component, a final driving instruction for the variable frequency compressor and the electronic expansion valve is generated.

2. The temperature intelligent regulating system of an industrial water chiller according to claim 1, wherein, When the control unit detects that the pressure change gradient value is positive and exceeds the dynamic noise threshold value, it is identified that the system is in a load surge state; the original second control component comprises an instruction for increasing the opening degree of the electronic expansion valve by a first preset step within a preset time, and an instruction for increasing the rotating speed of the variable frequency compressor by a second preset step; the control unit is also provided with a weight attenuation rule, after the final driving instruction is generated, with the falling of the pressure change gradient value, the weight of the final second control component in the final driving instruction is gradually reduced by a preset time constant until it is zero.

3. The temperature intelligent regulating system of an industrial water chiller according to claim 1, wherein, The dynamic noise threshold value is not a fixed value, but is determined in real time by the control unit based on the current operating frequency of the variable frequency compressor; the control unit is internally stored with a reference noise spectrum, and the reference noise spectrum defines the pressure fluctuation base value caused by the inherent vibration of the variable frequency compressor at different operating frequencies; the control unit queries the reference noise spectrum in real time to update the dynamic noise threshold value.

4. The temperature intelligent regulating system of an industrial water chiller according to claim 1, wherein, During the generation of the final second control component, the control unit continuously monitors the change trend of the pressure change gradient; when the inflection point feature that the pressure change gradient changes from increasing to decreasing is detected, the control unit identifies the time of the inflection point feature as an energy balance point; the control unit performs a state locking operation at the energy balance point, sets the current refrigerating capacity determined by the final second control component as a reference refrigerating capacity for subsequent control, and terminates the further dynamic adjustment of the final second control component, and instead adjusts the reference refrigerating capacity based on the first control component.

5. The temperature intelligent regulating system of an industrial water chiller according to claim 2, wherein, The control unit is further configured to calculate a time variation rate of the operating frequency of the variable frequency compressor in real time; when the final second control component drives the variable frequency compressor to accelerate and the time variation rate exceeds a preset acceleration threshold, the control unit generates a third control component for the electronic expansion valve based on the time variation rate; The third control component is configured to drive the opening degree of the electronic expansion valve to increase at a rate synchronized with the acceleration of the variable frequency compressor, and the increasing amplitude is positively correlated with the time variation rate of the operating frequency of the variable frequency compressor, so as to compensate for the decrease in suction pressure caused by the suction effect of the variable frequency compressor before the evaporation rate of the refrigerant in the evaporator increases.

6. The temperature intelligent regulating system of an industrial water chiller according to claim 1, wherein, The system further comprises a second pressure acquisition unit configured to acquire the refrigerant pressure on the condensation side; the control unit internally stores a differential pressure gain correction model, which defines a nonlinear mapping relationship between the feedforward regulation gain and the system operating differential pressure; the control unit calculates the differential pressure between the refrigerant pressure on the condensation side and the refrigerant pressure on the suction side of the variable frequency compressor in real time, and determines a current differential pressure correction coefficient according to the differential pressure gain correction model. When the final second control component is generated, the control unit performs weighted scaling on the amplitude of the final second control component by using the differential pressure correction coefficient.

7. The temperature intelligent regulating system of an industrial water chiller according to claim 1, wherein, The control unit is configured to monitor the opening degree regulation instruction sent to the electronic expansion valve in real time, and calculate an estimated value of the suction pressure variation rate caused by the action of the electronic expansion valve based on a preset valve pressure response model; When the pressure variation gradient value is extracted, the control unit performs signal decoupling operation, subtracts the estimated value from the measured differential value of the refrigerant pressure, and obtains a net pressure gradient value representing the external heat load disturbance; when the absolute value of the net pressure gradient value exceeds a dynamic noise threshold, the control unit generates the original second control component.

8. The temperature intelligent regulating system of an industrial water chiller according to claim 5, wherein, The control unit is further configured to monitor the suction superheat degree on the suction side of the variable frequency compressor in real time, and calculate a time variation rate of the suction superheat degree; during the generation of the third control component for driving the opening degree of the electronic expansion valve to increase, the control unit performs logical verification: when it is detected that the time variation rate of the suction superheat degree is negative and the absolute value exceeds a preset safety threshold, a wet compression risk state is identified, and the control unit immediately shields or reverses the third control component, forces to maintain or reduce the current opening degree of the electronic expansion valve, until the time variation rate of the suction superheat degree returns to a safe range.

9. The temperature intelligent regulating system of an industrial water chiller according to claim 1, wherein, The control unit performs time domain differentiation processing on the refrigerant pressure by using a sliding window algorithm, and the time length of the sliding window is set to be less than the fluid transport delay time required for the cooling fluid in the refrigeration cycle loop to be transported from the heat source to the temperature acquisition unit.

10. The temperature intelligent regulating system of an industrial water chiller according to claim 1, wherein, Gain coefficient defined in the gain correction model with a safety margin satisfies the following relationship: wherein, is a preset warning value, is a low-voltage protection threshold, is a preset attenuation index greater than 1, is used to set the attenuation rate of the gain coefficient when approaching the low-voltage protection threshold.

Images