Peak-valley electricity price differentiated heat pump ambient temperature self-adaptive regulation system

Abstract

The application discloses a heat pump environment temperature self-adaptive regulation and control system oriented to peak-valley electricity price differentiation, relates to the technical field of environment temperature regulation and control, and comprises a disturbance identification module, a sampling scheduling module, a temperature correction module, a multi-source checking module, a regulation and control inhibition module and a self-optimization feedback module.The disturbance identification module constructs a temperature disturbance diagnosis baseline, extracts the change rate characteristics, spectrum mutation characteristics and temperature probe self-heating characteristics of a continuous external environment temperature sequence, generates a sampling risk index, and forms a rapid oscillation mark.Through the self-adaptive sampling, probe self-heating compensation, multi-source consistency checking and feedback optimization mechanism, the heat pump system can realize dynamic response to environment changes, improve data accuracy and control decision reliability, and significantly enhance the stability, energy saving and comfort control ability under complex working conditions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ambient temperature regulation, and particularly relates to a heat pump ambient temperature self-adaptive regulation system oriented to peak-valley electricity price differentiation. BACKGROUND

[0002] The heat pump ambient temperature self-adaptive regulation system oriented to peak-valley electricity price differentiation refers to an intelligent ambient temperature control device and method combining the peak-valley electricity price law of the power market and the operation characteristics of the heat pump. The core idea is that when the electricity price is in the low valley period, the system pre-regulates the building or ambient temperature to the set comfort interval or energy storage interval by increasing the heat pump working load, so as to reduce or avoid high-load operation in the high electricity price period; when the electricity price is in the peak period, the system dynamically reduces the heat pump operation power or delays operation according to the ambient temperature change and indoor-outdoor heat balance state, so as to reduce the energy consumption. At the same time, the system has the ambient temperature self-adaptive regulation capability, can collect the outdoor climate conditions, indoor comfort index and heat pump operation parameters in real time, optimize the operation strategy through the adaptive algorithm, and realize the fine control of the heating and refrigeration process. The method not only effectively reduces the peak and fills the valley, improves the utilization efficiency of power resources, but also significantly reduces the operation cost of users and improves the environmental comfort, which embodies the fusion characteristics of intelligent energy management and building energy-saving control.

[0003] The prior art has the following disadvantages:

[0004] In the prior art, the heat pump ambient temperature self-adaptive regulation process generally relies on fixed sampling periods to monitor the external temperature. However, in the dynamic scene of rapid oscillation of external temperature, if the sampling period cannot be updated adaptively according to the temperature change rate, the sensor will be in a high-frequency collection state for a long time. Since the prior art lacks a dynamic balance mechanism between the sensor working load and the sampling frequency, the temperature measurement probe is prone to overload drift in the long-time high-frequency operation process, so that the temperature measurement data gradually deviates from the true value. When the distorted temperature data is continuously fed back to the regulation logic, the operation strategy of the system will be seriously disturbed, resulting in deviation of the global operation decision, which is manifested as incorrect judgment of the ambient temperature trend. In a more serious case, the system will produce a regulation behavior completely opposite to the actual demand direction, such as incorrectly executing continuous heating operation when the ambient temperature rises sharply, or incorrectly reducing the heating power when the temperature drops sharply, so that the ambient temperature deviates greatly from the target interval, which not only destroys the comfort, but also causes abnormal energy consumption and potential damage risk of equipment.

[0005] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] The purpose of the present application is to provide a peak-valley electricity price differentiated heat pump environment temperature adaptive regulation system to solve the problems in the background art.

[0007] In order to achieve the above-mentioned purpose, the present application provides the following technical solution: a peak-valley electricity price differentiated heat pump environment temperature adaptive regulation system, comprising a disturbance identification module, a sampling scheduling module, a temperature correction module, a multi-source checking module, a regulation suppression module and a self-optimization feedback module:

[0008] The disturbance identification module constructs a temperature disturbance diagnosis baseline, extracts the change rate feature, spectrum mutation feature and temperature probe self-heating feature of the continuous external environment temperature sequence, generates a sampling risk index, and forms a rapid oscillation marker;

[0009] The sampling scheduling module executes an adaptive sampling strategy based on information gain according to the residual error between the rapid oscillation marker and the external temperature prediction trend and the current observation value, sets the upper and lower limits of the sampling period, shortens the period when the sampling risk index exceeds the threshold, lengthens the period when the temperature is stable, and outputs the probe load factor at the same time;

[0010] The temperature correction module estimates the probe self-heating temperature rise based on the probe load factor, combines the measurement current, air flow speed and heat conduction path parameters, deducts it from the collected data, and generates a corrected temperature sequence;

[0011] The multi-source checking module constructs a state estimation model based on the corrected temperature sequence, judges the consistency of multi-source observation, adjusts the sampling period and triggers the probe calibration when the deviation is out of bounds, and outputs the consistency result;

[0012] The regulation suppression module executes reverse regulation suppression control according to the consistency result, sets the peak-valley switching buffer time window, regulation integral threshold and output slope limit, executes order-preserving amplitude reduction when there is trend misjudgment risk, and feeds back the control result to the temperature disturbance diagnosis baseline;

[0013] The self-optimization feedback module writes the running residual error, energy consumption deviation, comfort deviation value and sensor overload number after reverse regulation suppression into the temperature disturbance diagnosis baseline, updates the threshold mapping, control gain and sampling period upper and lower limits, and realizes closed-loop control of disturbance identification, risk assessment, strategy adjustment and feedback optimization.

[0014] Preferably, the rapid oscillation marker formation step is as follows:

[0015] The temperature sensor installed on the outside of the building collects the environment temperature sequence at fixed time intervals, extracts the temperature change rate per unit time, forms a change rate sequence, and calculates the standard deviation of the change rate sequence in every 10-minute time window to determine the temperature change trend;

[0016] The continuous temperature sequence is analyzed in frequency by using a sliding time window, the amplitude ratio change of the medium and high frequency components is extracted, and it is judged whether there is a spectrum mutation behavior, and a spectrum mutation feature is formed.

[0017] Real-time measurement of sensor current and voltage, combined with air flow rate and probe heat conduction path parameters, estimate the self-heating temperature rise per unit time, and form the self-heating temperature rise feature;

[0018] The obtained features are weighted according to the set weight, and the sampling risk index is generated, and when the sampling risk index exceeds the set threshold, the rapid oscillation marker signal is output as the input basis for the next step of sampling period adjustment.

[0019] Preferably, the probe load factor generation step is as follows:

[0020] A linear prediction model is established based on the environmental temperature data in the previous time window, and the current actual collected temperature is compared with the predicted value to calculate the temperature prediction residual;

[0021] The temperature prediction residual and the rapid oscillation marker output in the previous step are jointly determined to determine the external temperature disturbance risk, and the necessity of sampling period adjustment is determined based on the information contribution degree;

[0022] According to the determination result, the sampling period is dynamically adjusted, the period is shortened in the high disturbance state, and the period is lengthened in the temperature stable state, and the temperature rise intensity is estimated by combining the sensor electric power, air flow rate and installation heat conduction path parameters to construct the probe load factor;

[0023] The prediction residual, rapid oscillation marker, adjusted sampling period and probe load factor are input into the next step of self-heating deviation compensation process to ensure the accuracy of probe temperature rise correction.

[0024] Preferably, the generation of the corrected temperature sequence step is as follows:

[0025] The output probe load factor is combined, and the current value, voltage value and duration in each sampling period are recorded to calculate the total amount of electric energy input per unit time;

[0026] The air flow rate difference between the upstream and downstream of the probe is measured to determine the local convective heat dissipation capacity, and the heat conduction path parameters in the installation structure are called to obtain the heat conduction material type, thickness, filler and thermal conductivity;

[0027] The self-heating temperature rise value of the current period is calculated according to the difference between the electric energy input and the heat dissipation capacity, and the temperature rise value is deducted from the original temperature value of the current period in the data processing link to form the corrected temperature collection data;

[0028] The corrected temperature data is transmitted to the next step of the multi-source observation consistency verification process, and if the temperature rise exceeds the limit for consecutive periods, the sampling period is automatically extended to reduce the probe load.

[0029] Preferably, the consistency result output step is as follows:

[0030] In each sampling period, the corrected outdoor temperature, heat pump supply air temperature and heat pump return air temperature are synchronously acquired, and the sampling time deviation is ensured to be no more than 2 seconds;

[0031] Based on the heat pump operation principle, a state judgment model is established to calculate the temperature difference between the supply air and return air and the temperature difference between the supply air and outdoor temperature, and compare them with the preset physical interval to determine whether there is data anomaly;

[0032] When the temperature difference result exceeds the preset interval, the current sampling period is shortened, and the temperature measurement probe automatic calibration process of the corresponding measurement point is triggered to improve the stability and reliability of temperature observation;

[0033] The observation consistency discrimination result of the current sampling period is output as the basis for control strategy adjustment, energy consumption evaluation and sampling strategy optimization.

[0034] Preferably, after completing the multi-source temperature observation consistency discrimination, the step of further executing reverse regulation suppression control includes:

[0035] Determine whether the current is in the peak-valley electricity price switching sensitive period, and set the running buffer time in the corresponding direction to delay the control response within 5 minutes before and after the electricity price switching is determined;

[0036] During the buffer period, determine whether the current temperature error exceeds the reverse regulation threshold, if not, freeze the cumulative output value of the integral controller, and set a static dead zone around the threshold to prevent frequent switching;

[0037] Under the condition of allowing regulation, limit the change slope of the control output power, and set the maximum power change amplitude per unit time to avoid instantaneous over-regulation;

[0038] In the state where the probe load factor is high and the sampling risk index continues to rise, the order-preserving amplitude reduction strategy is executed to reduce the regulation amplitude and set an observation period, and after completion, the control execution result is fed back to the temperature disturbance diagnosis baseline.

[0039] Preferably, after completing the reverse regulation suppression control, parameter self-adaptation update is executed, and the specific steps are as follows:

[0040] Collect the running residual, energy consumption deviation, environmental comfort deviation value and sensor overload trigger times, and organize them into a structured data frame;

[0041] The data frame is written into a temperature disturbance diagnosis structure, historical trends are analyzed based on a sliding window, and disturbance identification thresholds, control gain parameters and upper and lower limits of a sampling period are dynamically corrected;

[0042] The updated parameters are immediately applied to disturbance identification, control decision and data acquisition processes in the next control cycle, so as to realize adaptive adjustment of the operation strategy.

[0043] The parameter update content and trigger basis of each round are recorded, a parameter control structure with a version number and a time stamp is established, and is used for traceability analysis and subsequent optimization iteration.

[0044] In the above technical solution, the technical effects and advantages provided by the present application are as follows:

[0045] The present application automatically adjusts the sampling period according to the degree of environmental change, prevents the temperature measurement probe from producing drift errors due to overload, and simultaneously realizes self-heating compensation through the probe load factor to ensure the authenticity and consistency of the collected data. Further, through multi-source observation checking and trend discrimination mechanism, the control direction error caused by temperature misjudgment is effectively inhibited, and in the process of system operation, the control parameters are continuously optimized based on energy consumption, residual error and comfort feedback, so as to realize dynamic adaptive adjustment of the whole process from data acquisition, strategy decision to execution response. Compared with the traditional fixed strategy heat pump control method, the present application significantly improves the operation stability, energy saving effect and user comfort control level of the system under dynamic climate, price change and complex building thermal environment. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments or the prior art, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0047] Figure 1 The module schematic diagram of the heat pump environment temperature adaptive control system of the present application facing the peak-valley electricity price difference. DETAILED DESCRIPTION

[0048] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations, however, can be implemented in many different ways and should not be construed as limited to the examples set forth herein; rather, these examples are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example implementations to those skilled in the art.

[0049] The present application provides a method for adaptive control of heat pump environment temperature, comprising the following steps: Figure 1The peak-valley price difference-oriented heat pump ambient temperature adaptive regulation system shown comprises a disturbance identification module, a sampling scheduling module, a temperature correction module, a multi-source checking module, a regulation suppression module, and a self-optimization feedback module.

[0050] The disturbance identification module establishes a temperature disturbance diagnosis baseline, extracts a change rate feature, a frequency spectrum mutation feature, and a temperature probe self-heating feature based on a continuous external environment temperature sequence, generates a sampling risk index, and forms a rapid oscillation marker according to the sampling risk index.

[0051] To solve the problem of fixed sampling period, distorted temperature data, and frequent strategy misjudgment in heat pump environment regulation under the scenario of rapid external temperature fluctuation in the prior art, a disturbance diagnosis method is constructed with the behavior feature of the environment temperature as the core, the temperature change trend is structurally analyzed, the high-risk disturbance state is identified in advance, and criterion support is provided for subsequent sampling frequency dynamic adjustment and temperature data correction. This method relies on the time sequence feature, frequency feature, and temperature probe operation characteristic of the external temperature sequence to generate a sampling risk evaluation mechanism with dynamic feedback capability, thereby realizing the feedforward optimization of the regulation strategy. The specific implementation process is as follows:

[0052] The temperature sensor configured on the outside of the building collects external environment temperature data, and the sampling frequency is initially set to 10 seconds once to obtain a temperature change sequence with continuity and representativeness. The collected temperature data needs to contain a timestamp identifier to ensure the consistency of the time interval in subsequent analysis. After obtaining data at not less than 30 time points, the temperature difference between every two adjacent sampling points is calculated, and the temperature change rate per unit time is calculated accordingly. By forming a continuous change rate sequence, the rising or falling trend of the temperature in the local time period can be obtained. Then, in every 10-minute time window, the maximum value, minimum value, mean value, and standard deviation of the change rate sequence are calculated to identify the severity of temperature fluctuation. To improve the accuracy of the judgment, the temperature change rate standard deviation threshold is set to 0.25℃ / min. If the standard deviation in the current time window exceeds the threshold, it indicates that there is obvious fluctuation behavior, which can be confirmed as an effective disturbance. This change rate feature is an important basis for judging temperature stability and will be one of the core parameters for sampling risk judgment in the subsequent steps.

[0053] After the above change rate feature extraction is completed, the temperature data of the same time period is subjected to frequency analysis. Specifically, the current temperature sequence is taken as input, and 30 minutes of continuous temperature data is selected by using a sliding time window method, and is slid forward by 1 minute as a step, and the temperature frequency component in each time window is calculated. The frequency analysis process obtains the amplitude distribution of different frequency components by mapping the temperature data to the frequency domain. In this process, the medium-high frequency components between 1 / 10 minutes and 1 / 1 minute are focused on, and the amplitude proportion change is recorded. If the relative growth of the medium-high frequency amplitude component exceeds 30% in two consecutive time windows, and the overall energy distribution moves up (i.e., the frequency peak shifts to the high frequency region), it can be determined that there is a frequency spectrum mutation phenomenon. This mutation means that there is rapid air mass movement, sudden sunlight, artificial intervention, and other nonlinear factors in the external environment, which is a significant signal of temperature rapid oscillation. The extraction result of the frequency spectrum mutation feature and the aforementioned change rate feature jointly constitute a two-dimensional description of the temperature disturbance behavior, providing complementary information support for subsequent comprehensive judgment.

[0054] After the change trend and frequency spectrum fluctuation features are identified, the influence of the self-heating behavior of the temperature measuring probe in the continuous sampling state on the temperature measurement accuracy is further considered. To this end, a current detection unit and a voltage detection circuit are added in the sampling circuit to monitor the working current and voltage of the sensor in real time, and the air flow speed around the probe is measured by a wind speed sensor in the ventilation duct to obtain the heat exchange rate. At the same time, the installation position material type (such as metal pipe, plastic shell, bare installation, etc.) and its thickness of the temperature measuring probe are labeled, and the thermal conductivity parameters are queried. According to the measured current, voltage and time, the power consumption of the probe per unit time is calculated, and the heat dissipation capacity is calculated combined with the installation structure and ventilation speed. Through energy conservation analysis, the temperature rise value that the probe may generate per unit time, i.e., the self-heating temperature rise deviation, is obtained. In implementation, if the calculated temperature rise exceeds 0.4°C, it is considered that the current working state of the probe may have a significant impact on the temperature measurement result. The quantization result of this self-heating feature will be included in the sampling risk assessment together with the change rate feature and the frequency spectrum feature to identify whether the current sampling data is reliable.

[0055] Based on the extraction results of the above three characteristics, that is, the temperature change rate feature, the spectrum mutation feature and the self-heating temperature rise feature, a sampling risk index is constructed. In order to realize comprehensive judgment, weights are set for the three features, for example, the change rate accounts for 40%, the spectrum mutation accounts for 35%, and the self-heating feature accounts for 25%. The numerical weights of the three features are weighted according to the set weights, and the calculation results are normalized to obtain a sampling risk index ranging from 0 to 1. The risk index threshold is set to 0.7, if the current index exceeds the threshold, it is determined to be in a high disturbance state, and a fast oscillation marker signal is output. The signal will be used as the front input of the subsequent sampling frequency dynamic adjustment, temperature data correction, control strategy buffering and safety limiting mechanism, and will run through the decision logic of the whole environmental regulation process.

[0056] The role of this step is to identify the trend of rapid change of external environment temperature and the risk of sampling distortion in advance by constructing a temperature disturbance diagnosis baseline, thereby providing criterion support for subsequent sampling frequency dynamic adjustment and temperature measurement data correction. By extracting the change rate feature, the spectrum mutation feature and the self-heating temperature rise feature of the temperature probe from the continuous temperature sequence, the temperature change behavior is described in multiple dimensions, which not only can identify the fast disturbance state in the natural environment, but also can effectively distinguish the false temperature drift caused by the heating of the sensor itself. By fusing the above features, a sampling risk index is formed, and a threshold is set to judge whether it is in a high disturbance state, thereby outputting a fast oscillation marker, providing a front input for the active response and fine control of the heat pump regulation system. This step not only enhances the system's perception ability to environmental dynamics, avoids the deviation of control strategy caused by false temperature data, but also provides a reliable guarantee for the stability and energy saving of the subsequent control logic, which is different from the extensive control mode in the prior art which relies on a single temperature threshold or fixed period response, and has stronger intelligence and robustness.

[0057] The sampling scheduling module executes an adaptive sampling strategy according to the residual error between the fast oscillation marker and the predicted trend of the external temperature and the current observed value, sets the upper and lower limits of the sampling period based on the information gain, shortens the sampling period when the sampling risk index exceeds the set threshold, lengthens the sampling period when the external environment temperature tends to be stable, and outputs the probe load factor;

[0058] After the construction of the temperature disturbance diagnosis baseline, multiple disturbance behavior characteristics including the temperature change rate feature, the spectrum mutation amplitude and the self-heating temperature rise of the temperature probe are obtained, and the sampling risk index is generated based on the above characteristics and the fast oscillation marker is output. In order to further improve the response accuracy of sampling regulation, a residual error comparison method based on the predicted trend and the actual sampling value is adopted to establish a sampling period adjustment method for disturbance dynamic identification and energy efficiency optimization target, and the temperature measurement behavior is realized. The specific steps are as follows:

[0059] With the past 30 minutes as a sliding time window, a continuous external environment temperature data sequence is extracted, and a series of temperature sequence samples of time periods are obtained by sliding forward every minute. In each time window, the least square method is used for linear fitting to obtain the slope and intercept of the trend line, which constitutes the temperature prediction model of the time period. Taking the current time as the benchmark, the prediction model of the previous time window is used to calculate the current temperature value, and the prediction value is compared with the actual temperature collected at the current time to calculate the temperature residual error. If the absolute value of the temperature residual error exceeds 0.3°C, it is considered that the actual temperature behavior deviates from the historical trend, indicating that there is a disturbance risk at the current time point.

[0060] The temperature prediction residual error and the fast oscillation marker calculated in the previous step are jointly determined. The fast oscillation marker reflects the state that the sampling risk index is higher than 0.7, and the temperature residual error reflects whether the change trend of the current temperature behavior is abnormal. If both conditions are met, i.e., there are historical disturbance characteristics and current prediction deviation, it is confirmed that the current state is a high-risk disturbance state, and the sampling behavior should respond quickly; otherwise, if the fast oscillation marker is in the non-activated state and the residual value is continuously within the stable range of 0.2°C, it is determined that the external environment temperature fluctuation is small, and the current sampling frequency can be appropriately reduced. In the joint determination, an information contribution evaluation method is further introduced, taking the contribution degree of the current sampling data to the control strategy as the standard to judge whether to continue to maintain the sampling period or to adjust it, so as to avoid increasing invalid calculation and probe load in the case of redundant sampling.

[0061] Based on the results of joint determination, the dynamic adjustment of the sampling period is performed. The upper and lower limits of the initial sampling period are set to 5 seconds and 60 seconds, respectively, and the sampling period is set to 15 seconds in the normal state. When in a high disturbance state and the current sampling period is higher than 10 seconds, the sampling period is directly shortened to 70% of the previous period; if it is continuously determined to be a temperature stable state for 3 minutes and the sampling period is less than 45 seconds, the sampling period is extended to 130% of the previous period. At the same time of adjusting the sampling period, the working current and voltage of the temperature measuring probe and the air flow velocity around the measuring point in the ventilation duct are monitored in real time. The current electric power input (by multiplying the current and voltage) and the air flow velocity are used to estimate the heat exchange strength of the sensor, and combined with the physical structure and installation material (such as copper guide sleeve, plastic card slot or aluminum heat dissipation base) thermal conductivity of the sensor, the sensor temperature rise amplitude per unit time is inversely calculated, and a probe load factor is constructed accordingly, which is used to reflect the influence strength of the sampling frequency adjustment on the probe thermal load.

[0062] The current temperature prediction residual, the fast oscillation mark state, the adjusted sampling period and the probe load factor are input as data together into the subsequent probe self-heating deviation compensation processing stage to support the estimation and correction accuracy improvement of the sensor self-heating temperature rise. The sampling control strategy not only realizes the active response of the temperature collection behavior to the disturbance trend, but also realizes the feedback modeling of the sampling behavior to the change of the thermal load by introducing the load factor, avoids the thermal drift error accumulation problem caused by the fixed sampling or single sampling strategy in the prior art, and significantly improves the effectiveness of data collection and the response accuracy of the control strategy.

[0063] The role of this step is to build a temperature sampling mechanism with dynamic response capability, so as to realize the rapid perception and accurate response to external temperature disturbance in the heat pump environment regulation process. By fusing the residual information of the fast oscillation mark and the external temperature prediction trend, this method can judge whether the current temperature change deviates from the historical trend in real time, and actively adjust the sampling period accordingly. When the external environment temperature fluctuates rapidly, the response frequency is improved by shortening the sampling period to capture more key feature points; in the stable stage of temperature change, the sampling period is lengthened to reduce invalid measurement and reduce the sensor thermal load. By setting the upper and lower limits of the sampling period and introducing information gain as the adjustment basis, this step effectively breaks the rigid restriction of the "fixed sampling period" in the traditional sampling strategy. At the same time, by outputting the "probe load factor" directly related to the sensor thermal load, the pre-quantitative basis is provided for the subsequent probe self-heating compensation step, ensuring that the measurement data still has credibility under high-frequency operation. Overall, this step ensures the real-time data while considering energy consumption and accuracy, lays a data foundation for accurate execution of subsequent regulation logic, and is one of the key links to realize the whole-process adaptive control.

[0064] The temperature correction module dynamically estimates the self-heating temperature rise of the temperature measurement probe based on the probe load factor, the measurement current, the air flow speed distribution and the installation heat conduction path parameters, and real-time deducts the estimated value from the original temperature collection data to generate a corrected temperature sequence;

[0065] After the completion of the adaptive sampling period adjustment based on the fast oscillation flag and the temperature prediction residual, the system obtains the probe load factor within each sampling period, which reflects the working intensity of the temperature probe in the current operating state. However, relying solely on sampling frequency adjustment cannot solve the problem of sensor heat accumulation caused by high-frequency sampling, especially in the installation environment where the probe is poorly ventilated, the heat dissipation channel is limited, or the structural thermal conductivity is low. The heat energy is retained around the probe, which is easy to cause temperature rise deviation phenomenon, thereby affecting the accuracy of temperature data. To eliminate the measurement error caused by the self-heating of the probe, a method based on the actual working parameter estimation of the self-heating temperature rise and the data correction is proposed. The probe temperature rise is dynamically estimated, and the temperature measurement result is deducted in the data layer to correct the original collected data. The specific implementation process is as follows:

[0066] At the end of each sampling period, the probe load factor of the corresponding period is obtained. The factor is generated by the previous implementation step, and is calculated based on the number of samples per unit time, the current-on ratio, the power supply fluctuation, and the heat capacity parameters of the probe material. The value range is 0 to 1, and the higher the value represents the greater the thermal load of the probe. At the same time of obtaining the value, the system records the actual power supply current value, voltage value and sampling duration of the probe in this period. These parameters are obtained by real-time feedback from the sampling circuit: the current is calculated by the voltage drop after the series sampling resistor, the voltage is obtained by directly measuring the output potential of the stable power supply, and the sampling duration is recorded and output by the timer inside the microcontroller.

[0067] Record the air flow velocity distribution and the heat conduction path information of the probe installation structure. The air flow velocity is obtained by arranging a hot film anemometer on both sides of the probe upstream and downstream. The sensor records the instantaneous wind speed values at both positions at the start and end of each sampling period, with the unit being meters per second. The wind speed difference reflects the local air flow state. If the difference is less than 0.5 meters per second and the absolute wind speed is less than 1 meter per second, it means that the local air convection is weak and the heat exchange efficiency is low. In addition, the structural heat conduction path parameters are obtained by preset input or field calibration of the probe installation structure, including the sensor shell material (such as copper, stainless steel or aluminum alloy), the shell thickness (in millimeters), the internal filling material type (such as silicone, epoxy resin), and the area and contact pressure of the contact interface. The thermal conductivity parameters of each material are recorded by the installer during installation and are fixed in the device parameter table, which is saved in table form for real-time calling.

[0068] The self-heating temperature rise of the probe is calculated using the parameters obtained above. In the current cycle, the thermal energy input of the probe per unit time is equal to the product of the current and voltage, multiplied by the sampling time length, to obtain the total energy input. Assuming that all the electrical energy is converted into heat, in the case of low air flow or low structural heat conduction efficiency, the heat lingers on the surface of the probe, forming a temperature rise. To evaluate this temperature rise effect, the following factors need to be considered: whether the local wind speed can carry away the heat (convective heat dissipation ability), and the efficiency of heat conduction through the probe material to the outside (heat conduction path). The specific heat capacity and thermal conductivity of each material are obtained by looking up the table, and the heat release capacity under actual conditions is calculated. When the heat dissipation capacity is not enough to cover the heat energy input, the difference is converted into an equivalent temperature rise, which is the estimated value of the self-heating temperature rise in this cycle. The estimated value is expressed in degrees Celsius, with a precision of 0.01°C, which is used for subsequent data correction processing.

[0069] The estimated self-heating temperature rise value is deducted from the original temperature value collected in the current cycle in real time, forming a corrected temperature value. The deduction operation is performed in the data processing unit of the microcontroller, and the correction program is triggered immediately after each temperature sampling. If the estimated temperature rise in the current cycle is 0.48°C and the original temperature value is 26.35°C, the corrected value is 25.87°C. This value will be passed to the subsequent control strategy judgment logic for judging indoor and outdoor temperature difference, control strategy switching, and energy consumption evaluation, etc. To ensure data continuity and correction smoothness, a first-order filtering process is performed after the temperature rise is deducted to prevent system control fluctuations caused by single-cycle mutations. If the self-heating temperature rise exceeds 0.5°C for three consecutive cycles and the load factor remains above 0.9, the device will trigger a high-load reminder, automatically extending the sampling period by 10 seconds through a pre-set mechanism to reduce the probe load and prevent long-term temperature drift.

[0070] The purpose of this step is to identify and eliminate the temperature measurement error caused by the self-heating effect of the temperature probe during high-load operation, ensuring that the collected environmental temperature data is true and accurate. Through the current, voltage, sampling time, air flow speed distribution, and heat conduction characteristics of the probe installation structure, the possible temperature rise deviation of the probe under the current operating conditions is evaluated. Based on the quantitative value of the probe load factor, the temperature drift in this cycle is dynamically calculated, and the temperature rise value is deducted from the original temperature data in real time to obtain the corrected temperature sequence. This correction process can effectively avoid data drift caused by the heating of the probe itself, especially in high-frequency sampling, high-temperature environments, or poor heat dissipation conditions. Compared with traditional methods that rely on static calibration or fixed offset compensation, this step has the advantages of real-time and adaptability, and can continuously ensure the credibility of temperature data in dynamic operating scenarios, providing a solid data foundation for subsequent environmental control decisions and control accuracy. It is an indispensable key link in the process of building an intelligent temperature control system.

[0071] The multi-source checking module, on the basis of the corrected temperature sequence, constructs a state estimation model combining the return air temperature, supply air temperature and outdoor temperature to determine the consistency of multi-source observation. When the observation deviation exceeds the set range, the sampling period is adjusted and the temperature probe calibration process is triggered, and the consistency discrimination result is outputted;

[0072] After the self-heating correction of the original temperature collection data is completed, the system obtains the environmental temperature sequence processed by real-time compensation. However, a single measuring point may still be affected by external disturbances, local microclimate of installation position, long-term drift or aging of components, etc., causing unstable or distorted measurement results. To further improve the accuracy and stability of temperature observation, multi-source temperature observation data from the supply air outlet, return air outlet and outdoor end are introduced on the basis of the corrected temperature to construct a state judgment model with clear physical correlation to check the consistency of multi-source temperature data. This method can effectively identify the source of data anomalies and adjust the sampling frequency and start the hardware calibration process when the observation is inconsistent, to improve the real perception ability of the system to environmental state changes. The specific implementation process is as follows:

[0073] In each sampling period, the temperature values from three different positions are collected respectively. The first is the corrected outdoor environmental temperature data, which comes from the temperature probe installed on the outside of the building in a light-avoiding and ventilated position, and the self-heating deviation caused by current driving is deducted through the foregoing steps. The second is the outlet temperature value of the heat pump supply air outlet, which is measured by a thermistor temperature sensor installed at the outlet of the air duct, representing the heat exchange effect after the heat pump runs. The third is the inlet temperature value of the heat pump return air outlet, which is collected by a temperature sensor arranged at the inlet of the air duct near the indoor return air outlet, reflecting the current indoor thermal environment state. In order to ensure the time consistency of the three temperature points, all sampling operations need to be executed by the main controller, and the maximum time deviation is not more than 2 seconds, and the sampling time stamp is bound in the controller.

[0074] A physical state judgment model based on the operating principle of heat pump is established to analyze the logical relationship among three temperatures. In the winter heating operation condition, the return air temperature should be significantly higher than the outdoor temperature, the supply air temperature should be slightly higher than the return air temperature, and the temperature difference between the supply air and the outdoor temperature is usually between 15 and 35 degrees Celsius; in the summer cooling condition, the supply air temperature should be lower than the return air temperature, and the return air temperature is higher than the outdoor temperature. In each sampling period, the following two temperature difference indicators are calculated in turn: one is the temperature difference ΔT1 between the supply air temperature and the return air temperature; the other is the temperature difference ΔT2 between the supply air temperature and the outdoor temperature. ΔT1 and ΔT2 are compared with the reasonable temperature difference interval set in the season to determine whether there is an anomaly. If the value of ΔT1 is negative or less than 1 degree Celsius in the heating condition, it is considered that the supply air effect is abnormal; if ΔT2 exceeds 40 degrees Celsius or is less than 10 degrees Celsius, it is considered that the external environment temperature or the supply air data may have drift or distortion.

[0075] When any one of the two sets of temperature difference calculation results exceeds the preset physical reasonable range, two operations are performed: one is to adjust the current temperature sampling period, and the other is to trigger the calibration process of the related temperature measurement probe. The sampling period adjustment method is to shorten the current sampling period value by 50%, that is, to adjust the original 20-second period to 10 seconds, to enhance the data density in the case of short-time fluctuation and to improve the judgment accuracy. If two consecutive sampling periods are judged to be inconsistent, the sampling period can be further reduced to the minimum value, for example, 2 seconds. Then, according to the measurement point position corresponding to the abnormal judgment result, the automatic calibration process of the temperature measurement probe at this position is triggered. The automatic calibration process takes the factory calibration value of the probe as the reference, compares the output value with the real-time acquisition data of the standby redundant sensor in the return air / supply air channel, determines whether the current sensor has drift, and corrects the deviation value. In the calibration process, the sliding average smoothing method is used to buffer the output value to avoid the sharp fluctuation of the temperature control system caused by calibration jump.

[0076] After the above consistency judgment and strategy response are completed, the consistency discrimination result of the current sampling period is formed and output to the data processing channel as the precondition for subsequent control strategy switching and energy consumption optimization judgment. The consistency discrimination result is represented by a discrete value, which is divided into three types: the first type is "complete consistency", indicating that the thermodynamic relationship between all observation channels meets the model prediction and there is no data anomaly; the second type is "critical inconsistency", indicating that there is a slight out-of-bound but not enough to trigger the calibration process, only observation tracking is needed; the third type is "significant inconsistency", indicating that the basic logical relationship in the heat exchange model has been violated, and data correction, sampling frequency improvement and probe calibration need to be performed immediately. The discrimination result is cached in the controller for 3 cycles to analyze the stability trend of the current temperature monitoring system.

[0077] The role of this step is to build a state judgment model based on heat exchange relationship by introducing three independent sources of temperature observation data: return air temperature, supply air temperature and outdoor temperature, to check the physical consistency between multi-source temperature data, to identify potential temperature measurement errors or sensor drift problems. By analyzing whether the temperature difference between supply air and return air, and the temperature difference between supply air and outdoor, conforms to the heat pump operation logic, when inconsistencies or thermodynamic relationship violations are found, the system can automatically shorten the sampling period, increase the data update frequency, and start calibration operation on the possible abnormal sensors, to correct the measurement bias in time. This process ensures the integrity and reliability of temperature data, preventing global control strategy deviation caused by individual sensor errors. At the same time, the system outputs consistency discrimination results, providing reliable input for the next control decision. Compared with the single-channel dependent temperature control method in existing technology, this method improves the anti-interference ability and robustness of temperature collection, effectively supporting fine energy-saving control of heat pump systems in complex and dynamic environments.

[0078] The control inhibition module executes reverse regulation inhibition control according to the consistency discrimination result, sets a running buffer time window at the peak-valley electricity price switching time, sets a reverse regulation threshold in the control integral link, sets a slope limit in the power output link, executes a sequence-preserving amplitude reduction strategy when it is judged that there is a risk of operation trend misjudgment, and transmits the control execution result to the temperature disturbance diagnosis baseline;

[0079] In the previous implementation step, the consistency discrimination of multi-source observation based on return air temperature, supply air temperature and outdoor temperature is completed, and the result of judging whether the current temperature measurement data is real and reliable is obtained. In the case of "critical inconsistency" or "significant inconsistency" in the consistency discrimination result, it indicates that the current observation state may have deviation risk, and if the current temperature data is directly used as control input, it may cause misjudgment of control direction or strength, especially in the peak-valley electricity price switching sensitive period, which is more likely to cause energy consumption fluctuation and system response disorder. To solve this problem, a reverse regulation inhibition method is proposed, which dynamically constrains the control strategy execution based on the consistency discrimination result, fine controls from four dimensions of time delay, control integral, power change slope and amplitude response, and feeds back the result to the temperature disturbance diagnosis baseline. The specific implementation process is as follows:

[0080] At the start of each control cycle, the consistency discrimination result of the previous cycle output is received. If the result is "significant inconsistency", the reverse adjustment inhibition mechanism is triggered. At this time, it is judged whether the current is in the peak-valley electricity price switching period. The judgment method is: read the system clock and compare it with the preset electricity price switching time (such as 7:00, 11:00, 17:00, 21:00 every day), if the absolute time difference between the current time and any switching time point is less than or equal to 300 seconds (i.e. 5 minutes), it is determined that it is in the peak-valley switching sensitive area. In order to prevent misjudgment from causing regulation jump, a running buffer window is set during this period: if it is converted from low valley to high peak, the control response is delayed for 180 seconds, and the current output power is kept unchanged; if it is converted from high peak to low valley, it is delayed for 120 seconds, and the current regulation direction is locked. This buffering action can effectively inhibit the regulation distortion caused by the superposition of electricity price disturbance and unstable temperature data.

[0081] During the control response has not been executed, the judgment process of the regulation integral element is entered. Taking heat pump heating as an example, the current indoor target temperature is set to 22 degrees Celsius, the return air temperature is actually 21.6 degrees Celsius, and the consistency discrimination is "critical inconsistency". The system first calculates the current error as 0.4 degrees Celsius, and judges whether it exceeds the set reverse adjustment threshold. The reverse adjustment threshold is 0.5 degrees Celsius, when the current temperature error does not exceed this value, the power boost or mode switching operation is not started, but the cumulative value of the integral controller is frozen, and the power is kept constant. The operation aims to block the misadjustment signal formed by the superposition of small amplitude error, and prevent the existence of data risk from causing false accumulation behavior. At the same time, when the threshold value is floating up and down, a static hysteresis interval of ±0.1 degrees Celsius is set for the controller output, so that it will not frequently switch state near the critical value.

[0082] Under the premise of allowing the controller output power to change, a power change slope limiting mechanism is further introduced. Specifically: the power adjustment instruction is limited to not more than a certain maximum change rate within a unit time. For example, the maximum rising rate is set to not more than 0.5 kW per minute, and the maximum falling rate is set to not more than 0.8 kW per minute. In actual execution, each time the controller output power instruction is executed, the power change amplitude is calculated by comparing it with the last cycle power, if it exceeds the set slope range, the control output is increased or decreased in multiple cycles. For example, if the last cycle power is 3.0 kW, and the calculated result of this cycle is 4.2 kW, which exceeds the maximum rising rate, only rise to 3.5 kW is executed, and the remaining part is adjusted in the next cycle. This process is completed in a linear segmented way, without relying on complex mathematical modeling, which is convenient for controller implementation and maintenance. By setting the power change slope limit, the system transient overload or regulation jitter caused by sensor data fluctuation can be avoided.

[0083] When it is determined that the current state is at risk of misjudgment, and the trend of regulation cannot be accurately determined, the order-preserving and amplitude-reducing control strategy is enabled. The order-preserving refers to actively limiting the regulation amplitude on the premise of maintaining the current regulation direction, and preferentially controlling the heat pump operation in a small step adjustment or maintaining the current state. For example, under the condition that the target temperature is 22 degrees Celsius, the current actual temperature is 21.5 degrees Celsius, and the original plan is to increase the power by 1 kilowatt, if the probe load factor is greater than 0.9, and the self-thermal temperature rise in the past two periods exceeds 0.6 degrees Celsius, and the sampling risk index is continuously higher than 0.75, then the power is only increased by 0.3 kilowatt, and the observation period is set to last for two sampling periods. If the consistency result improves during this period, the original adjustment amplitude is allowed to be restored. This strategy introduces the concept of regulation strength "amplitude limiting" to ensure the stability of the operation trend and reduce the risk of reverse regulation caused by misjudgment.

[0084] After performing all the above inhibition control actions, all key control parameters executed in the current period, including operation buffer time, reverse regulation threshold state, actual power adjustment value, order-preserving execution flag, and other information, are transmitted as a set of structured data into the temperature disturbance diagnosis baseline. These information will be used to update the risk judgment rules, adjust the sensitive weight of the sampling risk index, and dynamically optimize the coupling relationship between sampling scheduling and regulation response in subsequent periods, forming a closed-loop control process from risk perception to response control to feedback optimization.

[0085] The role of this step is to dynamically inhibit the response strength and adjustment direction in the heat pump regulation strategy when the system determines that the current temperature state is abnormal or at risk of trend misjudgment based on multi-source observation data, to avoid making incorrect control decisions based on unstable or distorted data. Specifically, by setting the operation buffer time of the peak-valley electricity price switching period, the error response caused by the superposition of electricity price changes and temperature misjudgment is prevented; by setting the reverse regulation threshold in the regulation integral link, over-regulation caused by small amplitude fluctuations or error accumulation is avoided; by introducing slope limitation at the power output end, the amplitude of power change per unit time is limited to reduce the system jitter risk; and by enabling the order-preserving and amplitude-reducing strategy when the trend misjudgment risk is identified, the current regulation direction is preferentially maintained, and only the output power is adjusted within the amplitude limit. Through these detailed inhibition strategies, the anti-interference ability of the system can be improved while maintaining the stability of the regulation, effectively preventing energy consumption abnormalities and comfort decline caused by data abnormalities, external disturbances, or electricity price changes. In addition, the execution results of this step are fed back to the disturbance diagnosis mechanism to provide a basis for subsequent sampling optimization and control gain adjustment, and to build a data-driven closed-loop self-optimization mechanism.

[0086] The self-optimization feedback module writes the running residual after reverse regulation inhibition, energy consumption deviation, environmental comfort deviation value and sensor overload trigger number into the temperature disturbance diagnosis baseline, dynamically updates the threshold mapping relationship, control gain parameter and sampling period upper and lower limit, and constitutes the closed-loop control process of disturbance identification, risk assessment, strategy adjustment and feedback optimization.

[0087] After completing the reverse regulation inhibition control in the previous implementation step, the system has actively corrected the adjustment behavior under the misjudgment risk condition, ensuring that the control strategy can be executed in a stable, safe and energy-saving manner in special cases such as inconsistent multi-source temperature data, price switching interference, and heavy sensor load. However, reverse regulation only belongs to the "emergency braking" mechanism, and if these processing results cannot be fed back to the adjustment of control parameters, true learning and optimization cannot be achieved. Therefore, it is necessary to record and feed back the key operating deviations and control response results in the current period to the temperature disturbance diagnosis process, and dynamically correct the core judgment parameters in the diagnosis model, so as to realize the self-adaptive evolution of the sampling strategy, control strength and risk discrimination mechanism. That is, a specific implementation path is provided for this purpose, which is specifically divided into the following steps:

[0088] After each control cycle is completed, the key performance indicators related to the operating results of the cycle are collected and structured according to categories. The indicators to be collected include: (1) running residual, which refers to the difference between the set target temperature and the actual indoor environment temperature, in degrees Celsius, used to reflect whether the control target is achieved; (2) energy consumption deviation, which refers to the difference between the actual device consumed energy and the theoretical predicted energy in the cycle, in kilowatt-hours, used to measure the control efficiency; (3) environmental comfort deviation value, which considers the offset between temperature fluctuation frequency, fluctuation amplitude and target temperature stable interval, and uses a dimensionless index between 0 and 1 to represent, where the higher the value, the greater the comfort deviation; (4) sensor overload trigger number, which records the number of sampling adjustment actions triggered by the temperature probe load factor exceeding 0.85 in the current period. All indicators are collected by the controller and structured into a structured data frame, providing a basis for subsequent parameter correction.

[0089] The above data is written into the preset operation feedback update channel of the temperature disturbance diagnosis structure, and the historical index trend of the last five cycles is summarized in the form of a sliding window. According to the operation residual trend, the disturbance identification threshold is corrected: if the absolute value of the operation residual exceeds 0.6 degrees Celsius for three consecutive cycles, it indicates that the existing disturbance detection sensitivity is insufficient, and the temperature change rate threshold will be adjusted from the original value of 0.2 degrees Celsius / minute to 0.15 degrees Celsius / minute, improving the identification ability of small fluctuations; otherwise, if the residual remains below 0.3 degrees Celsius and the comfort deviation value is small, the threshold can be increased to 0.25 degrees Celsius / minute to prevent false triggering of high-frequency sampling. According to the energy consumption deviation trend, the control gain parameter is fine-tuned: if the energy consumption deviation exceeds 8% continuously and the temperature control effect does not improve significantly, the heating power control gain will be reduced by 10% to reduce the energy consumption required for unit temperature difference adjustment. If the sensor overload times accumulate more than three times in the last five cycles, the lower limit of the sampling period will be extended from the original 2 seconds to 4 seconds, and the sampling risk index judgment standard will be adjusted from 0.7 to 0.75 to alleviate the probe load pressure. All correction operations are automatically completed in the microprocessor according to the preset logic, ensuring real-time and stability of the adjustment.

[0090] After completing the parameter update, the update results are immediately applied to the next cycle of the adjustment behavior generation process. The newly set disturbance identification threshold will be used to analyze whether the rate of change of the external temperature change curve meets the disturbance trigger condition; the control gain parameter will directly affect the power output decision logic to adjust the output intensity; the upper and lower limits of the sampling period are set to adjust the information collection frequency, forming a response mechanism of "high risk high frequency sampling, low risk low frequency sampling". For example, when the updated threshold in the diagnosis baseline increases the sensitivity of temperature change rate, it will identify the disturbance trend earlier under the same environmental change and execute the strategy response in advance; when the control gain is reduced, the output power change is limited under the same temperature difference input, avoiding abnormal energy consumption caused by the previous strategy. The redefinition of the upper and lower limits of the sampling period also makes the collection mechanism more suitable for the device health status and the degree of environmental disturbance, reducing the risk of probe aging and drift.

[0091] To ensure that the adjustment results are traceable and reproducible, all parameter values, execution time, and corresponding trigger indicators of this round of update are written into an independent version control structure, and parameter version numbers and time stamps are assigned. Version information includes but is not limited to: control gain values before and after updating, upper and lower limits of sampling period, disturbance threshold setting, trigger reason, key trigger data items, and actual control effect profile. In subsequent operation, if it is found that the control effect is decreased due to parameter update, it can be rolled back to the previous optimal state according to the version number. This structure also supports manual review and engineering intervention, facilitating engineering evaluation and revision suggestions during implementation.

[0092] The role of this step is to realize the closed-loop self-optimization control of the whole process of adaptive regulation of the heat pump environment temperature. By writing the key performance indicators such as operation residual error, energy consumption deviation, environmental comfort deviation value and sensor overload trigger number back to the temperature disturbance diagnosis structure after completing the reverse regulation inhibition control, the system can dynamically adjust the core decision parameters, including disturbance identification threshold, control gain coefficient and sampling period upper and lower limit, so that the subsequent sampling and regulation strategy is closer to the actual running state. This step introduces a periodic operation feedback mechanism, so that the system not only responds to the current disturbance, but also continuously corrects the strategy in long-term operation, strengthens the regulation accuracy and stability, and improves the energy saving effect and equipment life. Compared with the static parameter setting and fixed control logic in the prior art, this step has obvious adaptive learning ability, can actively adjust the control behavior according to the operation result, embodies the closed-loop intelligent control characteristics of the system "perception-judgment-response-correction", and is the core support mechanism to realize long-term efficient and stable operation.

[0093] The present application automatically adjusts the sampling period according to the degree of environmental change, prevents the temperature measurement probe from producing drift error due to overload, and realizes self-heat compensation through the probe load factor to ensure the authenticity and consistency of the collected data. Further, through the multi-source observation checking and trend discrimination mechanism, the control direction error caused by temperature misjudgment is effectively inhibited, and in the system operation process, the control parameters are continuously optimized based on energy consumption, residual error and comfort feedback, realizing dynamic adaptive adjustment of the whole process from data collection, strategy decision to execution response. Compared with the traditional fixed strategy heat pump control method, the present application significantly improves the operation stability, energy saving effect and user comfort control level of the system under dynamic climate, price change and complex building thermal environment.

[0094] The above only describes some exemplary embodiments of the present application by way of illustration, and it is needless to say that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present application. Therefore, the above figures and description are illustrative in nature and should not be understood as limiting the scope of protection of the claims of the present application.

Claims

1. A heat pump ambient temperature self-adaptive regulation system oriented to peak-valley electricity price differentiation, characterized in that, The disturbance identification module, the sampling scheduling module, the temperature correction module, the multi-source checking module, the regulation and control inhibition module, and the self-optimization feedback module are included. The disturbance identification module constructs a temperature disturbance diagnosis baseline, extracts the change rate characteristics, frequency spectrum mutation characteristics, and self-heating characteristics of the continuous external environment temperature sequence, generates a sampling risk index, and forms a rapid oscillation marker. The sampling scheduling module executes an adaptive sampling strategy based on information gain according to the residual error between the rapid oscillation marker and the external temperature prediction trend and the current observation value, sets the upper and lower limits of the sampling period, shortens the period when the sampling risk index exceeds the threshold, lengthens the period when the temperature is stable, and outputs the probe load factor. The temperature correction module estimates the probe self-heating temperature rise based on the probe load factor, combines the measurement current, air flow speed, and heat conduction path parameters, deducts the probe self-heating temperature rise from the collected data, and generates a corrected temperature sequence. The multi-source checking module constructs a state estimation model based on the corrected temperature sequence, joint return air temperature, supply air temperature, and outdoor temperature, judges the consistency of multi-source observation, adjusts the sampling period when the deviation exceeds the boundary, and triggers the probe calibration, and outputs the consistency result. The regulation and control inhibition module executes reverse adjustment and inhibition control according to the consistency result, sets the peak-valley switching buffer time window, regulation and control integral threshold, and output slope limit, executes order-preserving amplitude reduction when there is trend misjudgment risk, and feeds back the control result to the temperature disturbance diagnosis baseline. The self-optimization feedback module writes the running residual error after reverse adjustment and inhibition, energy consumption deviation, comfort deviation value, and sensor overload number into the temperature disturbance diagnosis baseline, updates the threshold mapping, control gain, and sampling period upper and lower limits, and realizes closed-loop control of disturbance identification, risk assessment, strategy adjustment, and feedback optimization.

2. The peak-valley electricity price differentiated heat pump ambient temperature adaptive regulation system according to claim 1, characterized in that, The rapid oscillation marker formation steps are as follows: The temperature sensor installed on the outside of the building collects the environment temperature sequence at fixed time intervals, extracts the temperature change rate per unit time, forms the change rate sequence, and calculates the standard deviation in each 10-minute time window to determine the temperature change trend. The continuous temperature sequence is analyzed using a sliding time window method to extract the amplitude proportion change of the medium-high frequency component and determine whether there is a frequency spectrum mutation behavior to form the frequency spectrum mutation characteristics. Real-time measurement of sensor current and voltage, combined with air flow speed and probe heat conduction path parameters, estimates the self-heating temperature rise per unit time to form the self-heating temperature rise characteristics. The obtained characteristics are weighted according to the set weight to generate a sampling risk index, and a rapid oscillation marker signal is output when the sampling risk index exceeds the set threshold, serving as the input basis for the next step of sampling period adjustment.

3. The peak-valley electricity price differentiated heat pump ambient temperature adaptive regulation system according to claim 2, characterized in that, The probe load factor generation steps are as follows: A linear prediction model is established based on the environment temperature data in the previous time window, and the actual collected temperature is compared with the predicted value to calculate the temperature prediction residual error. The temperature prediction residual error and the rapid oscillation marker output in the previous step are used to determine the external temperature disturbance risk, and the necessity of sampling period adjustment is determined based on the information contribution degree. According to the determination result, the sampling period is dynamically adjusted, the period is shortened in the high disturbance state, and the period is lengthened in the temperature stable state, and the sensor electric power, air flow speed and installation heat conduction path parameters are combined to estimate the temperature rise strength, and a probe load factor is constructed; The prediction residual, the fast oscillation marker, the adjusted sampling period and the probe load factor are input into the next step of the self-heating deviation compensation process to ensure the accuracy of the probe temperature rise correction.

4. The peak-valley electricity price differentiated heat pump ambient temperature adaptive regulation system according to claim 3, characterized in that, The corrected temperature sequence generation step is as follows: Combine the output probe load factor, and record the current value, voltage value and duration in each sampling period, calculate the total amount of electric energy input per unit time; Measure the air flow speed difference between the upstream and downstream of the probe, judge the local convection heat dissipation capacity, and call the preset heat conduction path parameters in the installation structure to obtain the heat conduction material type, thickness, filler and thermal conductivity; According to the difference between the electric energy input and the heat dissipation capacity, the self-heating temperature rise value of the current period is calculated, and the temperature rise value is deducted from the original temperature value of the current period in the data processing link to form the corrected temperature collection data; The corrected temperature data is input into the next step of the multi-source observation consistency checking process, and if the temperature rise exceeds the limit for continuous multiple periods, the sampling period is automatically lengthened to reduce the load of the probe.

5. The peak-valley electricity price differentiated heat pump ambient temperature adaptive regulation system according to claim 4, characterized in that, The consistency result output step is as follows: In each sampling period, the corrected outdoor temperature, heat pump supply air temperature and heat pump return air temperature are obtained synchronously, and the sampling time deviation is ensured to be not more than 2 seconds; Based on the heat pump operation principle, a state judgment model is established, the temperature difference between the supply air and the return air and the temperature difference between the supply air and the outdoor temperature are calculated respectively, and compared with the preset physical interval to judge whether there is data anomaly; When the temperature difference result exceeds the preset interval, the current sampling period is shortened, and the automatic calibration process of the temperature measuring probe of the corresponding measuring point is triggered to improve the stability and reliability of temperature observation; The observation consistency discrimination result of the current sampling period is output as the basis for control strategy adjustment, energy consumption evaluation and sampling strategy optimization.

6. The peak-valley electricity price differentiated heat pump ambient temperature adaptive regulation system according to claim 5, characterized in that, After completing the multi-source temperature observation consistency discrimination, the steps of further executing reverse regulation suppression control include: Determine whether the current is in the peak-valley electricity price switching sensitive period, and set the running buffer time in the corresponding direction to delay the control response within 5 minutes before and after the electricity price switching is judged; During the buffer period, it is judged whether the current temperature error exceeds the reverse regulation threshold value, if not, the cumulative output value of the integral controller is frozen, and a static dead zone is set near the threshold value to prevent frequent switching; Under the condition of allowing regulation, the change slope of the control output power is limited, the maximum power change amplitude per unit time is set to avoid instantaneous over-regulation; In the state that the probe load factor is high and the sampling risk index continues to rise, the order-preserving amplitude reduction strategy is executed to reduce the regulation amplitude and set an observation period, and after completion, the control execution result is fed back to the temperature disturbance diagnosis baseline.

7. The peak-valley electricity price differentiated heat pump ambient temperature adaptive regulation system according to claim 6, characterized in that, After completing the reverse regulation suppression control, the parameter self-adaptive update is executed, and the specific steps are as follows: Collect the running residual, energy consumption deviation, environmental comfort deviation value and sensor overload trigger times, and organize them into a structured data frame; The data frame is written into a temperature disturbance diagnosis structure, historical trends are analyzed based on a sliding window, and disturbance identification thresholds, control gain parameters, and upper and lower limits of a sampling period are dynamically corrected; The updated parameters are immediately applied to disturbance identification, control decision-making, and data acquisition processes in the next control cycle, enabling adaptive adjustment of the operation strategy; The parameter update content and trigger basis of each round are recorded, and a parameter control structure with version number and timestamp is established for traceability analysis and subsequent optimization iteration.

Images