A method and system for optimizing the load of an air source heat pump in a high-altitude environment

Abstract

This invention relates to the field of air source heat pump technology, specifically disclosing a method and system for optimizing air source heat pump load in high-altitude environments. The method involves collecting multi-source characteristic sequences; calculating the cumulative altitude deviation based on atmospheric pressure measurements and location altitude data to determine the blockage confidence of the measurement channel; when the blockage confidence exceeds a preset threshold, discarding the atmospheric pressure measurements and substituting the location altitude data, fan current change rate, and superheat dynamic response value into the altitude-pressure ratio mapping relationship to obtain an equivalent altitude estimate; calculating the load matching frequency rising envelope based on the equivalent altitude estimate and real-time pressure ratio measurement, and using Bayesian search to dynamically correct the heat demand tracking coefficient; using the corrected coefficient as the load optimization benchmark for the compressor frequency upper limit and frequency ramp rate, and monitoring pressure ratio fluctuations in real time to iteratively update the boundary coefficients of the mapping relationship; this invention can maintain compressor safety control and adaptive heat load matching even when sensors are blocked.

Description

Technical Field

[0001] This invention relates to the field of air source heat pump technology, and specifically to a method and system for optimizing air source heat pump load in high-altitude environments. Background Technology

[0002] When air source heat pumps are used in high-altitude areas, the low atmospheric pressure and low air density significantly reduce the heat exchange capacity on the evaporator side, leading to an increase in the compressor's suction specific volume, resulting in a higher pressure ratio and reduced heating performance. To adapt to high-altitude environments, existing technologies typically use atmospheric pressure sensors to measure ambient air pressure in real time and perform altitude compensation for compressor frequency control. For example, fan speed, electronic expansion valve opening, and the upper limit of compressor operating frequency are adjusted based on air pressure values ​​to suppress excessive pressure ratios.

[0003] The existing technology has the following shortcomings: In high-altitude, dusty environments, when atmospheric pressure sensors fail completely due to dust blockage and cannot provide accurate altitude information, existing technologies lack the ability to dynamically construct and iteratively correct the altitude-pressure ratio mapping relationship based solely on location altitude data, fan current change rate, and evaporator superheat dynamic response value. This would allow for the maintenance of safe compressor frequency control and adaptive matching of heat load even under extreme conditions where effective air pressure measurement is unavailable. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for optimizing the load of an air source heat pump in a high-altitude environment, so as to solve the problems mentioned above.

[0005] The objective of this invention can be achieved through the following technical solutions: A method for optimizing the load of an air-source heat pump in a high-altitude environment includes the following steps: S1: Collect real-time atmospheric pressure measurements, location altitude data, fan current change rate, and evaporator superheat dynamic response values ​​under high-altitude environment, and combine them into a multi-source feature sequence. S2: Calculate the cumulative altitude deviation based on the atmospheric pressure measurement and positioning altitude data in the multi-source feature sequence, and determine the blockage confidence of the measurement channel based on the cumulative altitude deviation. S3: When the blockage confidence exceeds the preset threshold, the atmospheric pressure measurement value is discarded, and the positioning altitude data, the fan current change rate and the superheat dynamic response value are substituted into the pre-constructed altitude-pressure ratio mapping relationship to obtain the equivalent altitude estimate. S4: Based on the equivalent altitude estimate and the real-time measured pressure ratio, calculate the rising envelope of the load matching frequency, and use Bayesian search to dynamically correct the heat demand tracking coefficient of the envelope. S5: The corrected heat demand tracking coefficient is used as the load optimization benchmark for the compressor frequency upper limit and frequency ramp rate, while the pressure ratio fluctuation is monitored in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship.

[0006] As a further aspect of the present invention: S2 specifically includes: The atmospheric pressure measurements within multiple consecutive sampling periods are converted into measured altitudes, and the instantaneous difference between the measured altitude and the location altitude data in each sampling period is calculated. Each instantaneous difference is assigned a weighting coefficient that decays exponentially over time and then accumulated to obtain the cumulative dynamic altitude deviation. The cumulative dynamic altitude deviation is compared with the pre-stored unblocked deviation range, and the normalized blockage confidence score is output.

[0007] As a further aspect of the present invention: the normalized blocking confidence of the output specifically includes: Determine the positional offset direction of the cumulative dynamic altitude deviation relative to the lower and upper limits of the unblocked deviation range; The corresponding mapping curve is selected based on the direction of position offset. The mapping curve outputs 0 confidence when the cumulative deviation is below the lower limit and outputs single-position confidence when it is above the upper limit. When the cumulative deviation is between the lower and upper limits, the confidence level is smoothly increased from 0 to single-location confidence level according to the ratio of the distance of the cumulative deviation from the lower limit to the interval width, thus obtaining the normalized blockage confidence level.

[0008] As a further aspect of the present invention: S3 specifically includes: Collect location altitude data, fan current change rate, superheat dynamic response value and corresponding measured pressure ratio of heat pump at different altitudes under non-clogging conditions to form multiple sets of mapping sample pairs; The ratio of the rising slope of the superheat dynamic response value to the change rate of the fan current in each mapping sample pair is extracted as an altitude-sensitive feature. The altitude-sensitive feature is then subjected to exponential fitting with the positioning altitude data to obtain the altitude-pressure ratio mapping relationship. Substitute the current location altitude data, the rate of change of wind turbine current, and the dynamic response value of superheat into the altitude-pressure ratio mapping relationship to calculate the corresponding equivalent altitude estimate.

[0009] As a further aspect of the present invention: the step of exponentially fitting the altitude-sensitive feature quantity with the positioning altitude data to obtain the altitude-pressure ratio mapping relationship specifically includes: Using location elevation data as the independent variable and elevation-sensitive features as the dependent variable, an initial exponential relationship is established, which includes undetermined base parameters and proportional parameters. According to the order of the positioning altitude data from smallest to largest, each sample pair is substituted in one by one to calculate the fitting deviation under the current exponential relationship. Based on the sign of the product of the fitting deviations of two adjacent sample points, the value direction of the base parameter is adjusted in reverse. When the sum of the absolute values ​​of the fitting deviations of all sample points no longer decreases after two consecutive iterations, the current base parameter and ratio parameter are fixed, and exponential fitting is completed to obtain the altitude-pressure ratio mapping relationship.

[0010] As a further aspect of the present invention: S4 specifically includes: The prior range of the heat demand tracking coefficient is determined based on the equivalent altitude estimate, and multiple discrete candidate coefficients are selected evenly within the prior range. Substitute each candidate coefficient into the rising envelope of the current frequency to predict the change in pressure ratio at the next moment, and calculate the deviation between the predicted value and the real-time measured value of pressure ratio. Based on the magnitude of the change deviation corresponding to each candidate coefficient, a posterior weight is assigned to each candidate coefficient. The candidate coefficient with the largest posterior weight is used as the corrected hot demand tracking coefficient. At the same time, the prior value range of the next search is narrowed with the hot demand tracking coefficient as the center.

[0011] As a further aspect of the present invention: the prediction of the pressure ratio change at the next moment specifically includes: Discretize the current frequency rising envelope into multiple rising steps according to time, with each step corresponding to a frequency increment. For each candidate coefficient, the frequency increment of each ascending step is scaled using the candidate coefficient to obtain the scaled step sequence, and the sequence is accumulated to predict the total frequency value at the next moment. Based on the differential relationship between the current real-time measured pressure ratio and the total frequency value, the pressure ratio change at the next moment is estimated using the forward differential approximation method.

[0012] As a further aspect of the present invention: S5 specifically includes: The real-time measured pressure ratio values ​​were continuously collected over multiple control cycles, and the magnitude and direction of change of the measured pressure ratio values ​​between two adjacent cycles were calculated. When the direction of change deviates from the pressure ratio change trend predicted by the current mapping relationship, the change amplitude is multiplied by a preset attenuation factor to obtain the adjustment step size of the boundary coefficient. By adjusting the step size in the same direction to increase or decrease the upper and lower boundary coefficients in the altitude-pressure ratio mapping relationship, the updated boundary coefficients can surround the extreme points of the current pressure ratio fluctuation.

[0013] As a further aspect of the present invention: the calculation process of the preset attenuation factor is as follows: Record the cumulative number of deviations between the current pressure ratio change direction and the predicted trend of the continuous deviation mapping relationship; The reciprocal of the cumulative number of divergences is used as the initial reference for the attenuation factor, and the initial reference is locked to the lower limit when the cumulative number of divergences exceeds the upper limit. The current change magnitude is compared with the exponentially weighted average of historical change magnitudes. The initial benchmark is adjusted based on the comparison results to obtain the final attenuation factor, so that the larger the change magnitude, the smaller the attenuation factor.

[0014] An air-source heat pump load optimization system for high-altitude environments includes: The multi-source feature acquisition module is used to collect real-time atmospheric pressure measurements, positioning altitude data, fan current change rate, and evaporator superheat dynamic response values ​​under high-altitude conditions, which are then combined into a multi-source feature sequence. The blockage confidence determination module calculates the cumulative altitude deviation based on the atmospheric pressure measurement value and the positioning altitude data in the multi-source feature sequence, and determines the blockage confidence of the measurement channel based on the cumulative altitude deviation. The equivalent altitude estimation module discards atmospheric pressure measurements when the blockage confidence exceeds a preset threshold. Instead, it substitutes the location altitude data, fan current change rate, and superheat dynamic response value into the pre-constructed altitude-pressure ratio mapping relationship to obtain the equivalent altitude estimation. The heat demand tracking coefficient correction module calculates the load matching frequency rising envelope based on the equivalent altitude estimate and the real-time measured pressure ratio, and uses Bayesian search to dynamically correct the heat demand tracking coefficient of the envelope. The load optimization benchmark control module uses the corrected heat demand tracking coefficient as the load optimization benchmark for the compressor frequency upper limit and frequency ramp rate, while monitoring the pressure ratio fluctuation in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship.

[0015] The beneficial effects of this invention are: (1) This invention uses the location altitude data, fan current change rate and superheat dynamic response value as redundant data sources, and constructs an altitude-pressure ratio mapping relationship based on exponential fitting. When the atmospheric pressure sensor is blocked due to the dusty environment of high-altitude mines, it can automatically discard the distorted measurement value and switch to the equivalent altitude estimate driven by multimodal data, thereby maintaining the continuity and stability of compressor frequency control, avoiding pressure ratio overshoot, compressor surge and protective shutdown caused by sensor blockage, and improving the operational reliability of air source heat pump in harsh dusty environments.

[0016] (2) This invention calculates the load matching frequency rising envelope based on the equivalent altitude estimate and the real-time measured pressure ratio, and uses Bayesian search to dynamically correct the heat demand tracking coefficient. At the same time, it monitors the pressure ratio fluctuation in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship, so that the compressor frequency upper limit and frequency rise rate can adaptively track the actual heat load changes. This avoids the problem of heat output being out of sync with demand under the traditional fixed frequency control strategy. Thus, even under sensor failure and severe altitude fluctuations, it can still ensure the stability of heat output and effectively maintain the nighttime heating demand of key areas such as workers' dormitories. Attached Figure Description

[0017] The invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1, please refer to Figure 1 As shown, this invention provides a method for optimizing the load of an air-source heat pump in a high-altitude environment, comprising the following steps: S1: Collect real-time atmospheric pressure measurements, location altitude data, fan current change rate, and evaporator superheat dynamic response values ​​under high-altitude environment, and combine them into a multi-source feature sequence. S2: Calculate the cumulative altitude deviation based on the atmospheric pressure measurement and positioning altitude data in the multi-source feature sequence, and determine the blockage confidence of the measurement channel based on the cumulative altitude deviation. S3: When the blockage confidence exceeds the preset threshold, the atmospheric pressure measurement value is discarded, and the positioning altitude data, the fan current change rate and the superheat dynamic response value are substituted into the pre-constructed altitude-pressure ratio mapping relationship to obtain the equivalent altitude estimate. S4: Based on the equivalent altitude estimate and the real-time measured pressure ratio, calculate the rising envelope of the load matching frequency, and use Bayesian search to dynamically correct the heat demand tracking coefficient of the envelope. S5: The corrected heat demand tracking coefficient is used as the load optimization benchmark for the compressor frequency upper limit and frequency ramp rate, while the pressure ratio fluctuation is monitored in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship.

[0021] Example 2: In S1, atmospheric pressure measurements, altitude data, fan current change rate, and evaporator superheat dynamic response values ​​under real-time high-altitude conditions are collected and combined into a multi-source feature sequence, specifically including: A digital atmospheric pressure sensor installed on the outdoor unit casing of the air source heat pump collects the atmospheric pressure measurement value of the current environment in real time with a sampling period of 1 second. The measurement accuracy of the atmospheric pressure sensor is not less than ±0.1 kPa.

[0022] Meanwhile, the GPS receiver module built into the heat pump controller acquires the device's location altitude data every 5 seconds, in meters. The fan current change rate is collected by connecting a Hall-effect DC current transformer in series in the outdoor fan power supply circuit. The fan operating current value is continuously read at 0.5-second intervals. The difference between the current value and the previous current value is divided by the sampling interval to obtain the current change rate in amperes per second.

[0023] The method for obtaining the dynamic response value of evaporator superheat is as follows: attach a platinum resistance temperature sensor to the inlet and outlet pipe walls of the evaporator, measure the refrigerant inlet temperature and outlet temperature respectively, subtract the inlet temperature from the outlet temperature to obtain the superheat, and then continuously record the superheat value for 3 sampling cycles. Take the maximum value of the superheat difference between adjacent cycles as the dynamic response value of superheat.

[0024] Finally, the atmospheric pressure measurement, positioning altitude data, fan current change rate, and superheat dynamic response value, which are aligned at the same time, are spliced ​​into a four-dimensional multi-source feature sequence according to a fixed data order and stored in the cache inside the controller.

[0025] Example 3: In S2, based on the atmospheric pressure measurements and positioning altitude data from the multi-source feature sequence, the cumulative altitude deviation is calculated, and the confidence level of the measurement channel blockage is determined based on the cumulative altitude deviation. Specifically, this includes: Atmospheric pressure measurements over multiple consecutive sampling periods are converted into measured altitudes, and the instantaneous difference between the measured altitude and the location altitude data in each sampling period is calculated.

[0026] A sliding window with a fixed length of 60 sampling periods is set inside the heat pump controller, with each sampling period lasting 2 seconds. For each sampling period within the window, the atmospheric pressure measurement value collected within that period is first read. Using the pressure-altitude conversion table pre-stored in the controller's read-only memory, the atmospheric pressure measurement value is converted into a measurement altitude in meters using linear interpolation.

[0027] Simultaneously, the positioning altitude data at the same moment is read. The measured altitude is subtracted from the positioning altitude data to obtain the instantaneous difference for that sampling period, with the unit of the difference still being meters. If the positioning altitude data is invalid due to signal loss, the current period is skipped and the system waits for the next valid data.

[0028] Each instantaneous difference is assigned a weighting coefficient that decays exponentially over time, and these coefficients are accumulated to obtain the cumulative dynamic altitude deviation. Specifically, for the 60 instantaneous differences within the sliding window, they are arranged in ascending order, with each instantaneous difference corresponding to a weighting coefficient. The weighting coefficient is calculated as follows: using the natural constant 2.71828 as the base, the exponent is the number of periods from the latest sampling period to the negative current instantaneous difference, divided by the decay time constant 20. The weighting coefficient for the latest sampling period (the 60th period) is 1.0, and the weighting coefficient for the oldest sampling period (the 1st period) is approximately 0.05.

[0029] Each instantaneous difference is multiplied by its corresponding exponential decay weight, and then all products are summed. The resulting sum is the cumulative dynamic altitude deviation. This cumulative deviation retains the main characteristics of recent deviations while gradually forgetting earlier deviations.

[0030] The cumulative dynamic altitude deviation is compared with the pre-stored unblocked deviation range, and the normalized blockage confidence score is output.

[0031] During the equipment's factory calibration phase, in a clean, dust-free laboratory environment with known altitude, 100 sets of dynamic altitude deviation cumulative values ​​were collected and calculated using the same sampling and calculation methods described above. The minimum value was taken as the lower limit of the non-clogging deviation interval, and the maximum value was taken as the upper limit. These lower and upper limits were pre-programmed into the controller's non-volatile memory. In actual operation, the currently calculated dynamic altitude deviation cumulative value was compared with this interval, and a dimensionless value between 0 and 1 was output as the clogging confidence level.

[0032] Determining the positional offset direction of the cumulative dynamic altitude deviation relative to the lower and upper limits of the uncongested deviation interval involves the following steps: First, reading the pre-stored lower and upper limits; then, comparing the current cumulative dynamic altitude deviation with each of the lower and upper limits. If the current cumulative amount is less than the lower limit, the positional offset direction is determined to be below the interval; if the current cumulative amount is greater than the upper limit, the positional offset direction is determined to be above the interval; if the current cumulative amount is between the lower and upper limits, the positional offset direction is determined to be within the interval.

[0033] The corresponding mapping curve is selected based on the direction of position offset. When the cumulative deviation is below the lower limit, the mapping curve outputs zero confidence; when it is above the upper limit, it outputs single-position confidence.

[0034] The controller has three preset mapping curves, corresponding to three scenarios: below the interval, within the interval, and above the interval. When the position offset is below the interval, a constant zero mapping curve is selected, and the blockage confidence is always output as 0, regardless of the difference between the current accumulated amount and the lower limit. When the position offset is above the interval, a constant one mapping curve is selected, and the blockage confidence is always output as 1. When the position offset is within the interval, a linearly increasing mapping curve is selected.

[0035] When the cumulative deviation is between the lower and upper limits, the normalized blockage confidence is obtained by smoothly increasing the confidence level from zero to single-location confidence level according to the ratio of the distance of the cumulative deviation from the lower limit to the interval width.

[0036] When the cumulative dynamic altitude deviation falls within the non-blockage deviation interval, the following steps are taken: First, calculate the distance between the current cumulative amount and the lower limit, which is the current cumulative amount minus the lower limit. Then, calculate the interval width, which is the upper limit minus the lower limit. Finally, divide the deviation distance by the interval width to obtain a ratio between 0 and 1. This ratio is directly used as the blockage confidence score. For example, if the current cumulative amount equals the lower limit, the ratio is 0, and the blockage confidence score is 0; if the current cumulative amount equals the upper limit, the ratio is 1, and the blockage confidence score is 1; if the current cumulative amount is at the midpoint of the interval, the ratio is 0.5, and the blockage confidence score is 0.5. This achieves a smooth increase in the blockage confidence score within the interval.

[0037] Example 4: In S3, when the blockage confidence exceeds a preset threshold, atmospheric pressure measurements are discarded. Instead, the location altitude data, fan current change rate, and superheat dynamic response value are substituted into a pre-constructed altitude-pressure ratio mapping relationship to obtain an equivalent altitude estimate. Specifically, this includes: The system collects location altitude data, fan current change rate, superheat dynamic response value, and corresponding measured pressure ratio of the heat pump at different altitudes under non-clogging conditions, forming multiple sets of mapping sample pairs.

[0038] Before the equipment leaves the factory, the air source heat pump is placed in an environmental test chamber that simulates different altitudes, ensuring that the air inlet of the atmospheric pressure sensor is unobstructed. Eight altitude gradients were set in the test chamber: 1000m, 1500m, 2000m, 2500m, 3000m, 3500m, 4000m, and 4500m. At each altitude gradient, after the heat pump stabilized, data was continuously recorded for 120 seconds. Every 2 seconds, a set of location altitude data, fan current change rate, superheat dynamic response value, and compressor measured pressure ratio were collected. The four data points at each sampling time were used as a mapping sample pair, resulting in 480 sample pairs. All sample pairs were stored in the controller's external memory for subsequent fitting calculations.

[0039] The ratio of the rising slope of the superheat dynamic response value to the rate of change of the fan current is extracted for each group of mapped samples and used as an altitude-sensitive feature. The altitude-sensitive feature is then subjected to exponential fitting with the positioning altitude data to obtain the altitude-pressure ratio mapping relationship.

[0040] For each pair of mapped samples, the rising slope is first extracted from the superheat dynamic response value: take the superheat values ​​for three consecutive sampling periods, subtract the value of the first period from the value of the third period, and divide by the time span of two sampling periods (i.e., 4 seconds) to obtain the rising slope in degrees Celsius per second. Then, divide the rising slope by the rate of change of the wind turbine current at the same moment (in amperes per second), and the quotient is the altitude-sensitive characteristic quantity, which is dimensionless.

[0041] Using location elevation data as the independent variable and elevation-sensitive features as the dependent variable, 480 sample pairs were substituted into an exponential fitting equation in order from low to high elevation. By iteratively adjusting the base and scaling parameters in the equation, the sum of squared deviations between the fitted curve and all sample points was minimized, thus completing the exponential fitting and establishing a mapping relationship between elevation and pressure ratio. This mapping relationship is then embedded in the controller in the form of a lookup table.

[0042] Substitute the current location altitude data, the rate of change of wind turbine current, and the dynamic response value of superheat into the altitude-pressure ratio mapping relationship to calculate the corresponding equivalent altitude estimate.

[0043] When the blockage confidence level exceeds a preset threshold of 0.7, the controller immediately stops using atmospheric pressure measurements. Then, it reads the current location altitude data, fan current change rate, and superheat dynamic response value, and calculates the current altitude-sensitive characteristic using the same method as in the second paragraph.

[0044] By substituting this characteristic quantity as the dependent variable into the established exponential mapping relationship and solving in reverse, the independent variable is obtained, yielding the corresponding equivalent altitude estimate. This equivalent altitude estimate is in meters and is used to replace the altitude information provided by the blocked atmospheric pressure sensor.

[0045] Using location elevation data as the independent variable and elevation-sensitive features as the dependent variable, an initial exponential relationship is established, which includes undetermined base parameters and proportional parameters.

[0046] At the start of exponential fitting, an initial relationship is established: the altitude-sensitive feature is equal to the scale parameter multiplied by the base parameter raised to the power of the positioning altitude data. Both the base and scale parameters are undetermined positive real numbers. The initial value of the base parameter is set to 1.01, and the initial value of the scale parameter is set to 0.001. This relationship describes the exponential trend of the altitude-sensitive feature as the positioning altitude data increases.

[0047] Following the order of the positioning altitude data from smallest to largest, each sample pair is substituted in one by one to calculate the fitting deviation under the current exponential relationship. Based on the sign of the product of the fitting deviations of two adjacent sample points, the direction of the base parameter is adjusted in reverse.

[0048] All 480 sample pairs were sorted in ascending order of their altitude data. First, using the current base parameter and scale parameter, the fitted value was calculated for the first sample pair (lowest altitude), and the difference between the fitted value and the measured altitude-sensitive feature was determined as the fitting deviation for that sample point. This calculation was then repeated for each subsequent sample point. After calculating the fitting deviations for every two adjacent sample points, they were multiplied. If the product was positive, it indicated that the two deviations had the same sign, and the adjustment direction of the base parameter remained unchanged. If the product was negative, it indicated that the deviations had different signs, and the adjustment direction of the base parameter was immediately reversed, i.e., from increasing to decreasing or from decreasing to increasing. The step size for each adjustment of the base parameter was one-thousandth of its current value.

[0049] When the sum of the absolute values ​​of the fitting deviations of all sample points no longer decreases after two consecutive iterations, the current base parameter and ratio parameter are fixed, and exponential fitting is completed to obtain the altitude-pressure ratio mapping relationship.

[0050] In a complete iteration, all 480 sample points are processed sequentially, and the absolute value of the fitting deviation for each sample point is calculated. These 480 absolute values ​​are then summed to obtain the total absolute value of the deviation for that iteration. This process is repeated multiple times, recording the total absolute value of the deviation after each iteration. The iteration stops when the total absolute value of the deviation is exactly equal in two consecutive iterations (i.e., it no longer decreases), and the base parameter and scale parameter used in the last iteration are taken as the final fixed parameters. Substituting these two parameters into the initial relational expression yields the final elevation-pressure ratio mapping relationship.

[0051] The controller stores the parameters of this mapping relationship in non-volatile memory for later use in equivalent altitude estimation.

[0052] Example 5: In S4, based on the equivalent altitude estimate and the real-time measured pressure ratio, the rising envelope of the load matching frequency is calculated, and the heat demand tracking coefficient of the envelope is dynamically corrected using a Bayesian search, specifically including: The prior value range of the heat demand tracking coefficient is determined based on the equivalent altitude estimate. Within this prior value range, multiple discrete candidate coefficients are selected evenly. Specifically, a heat demand tracking coefficient memory is set up inside the controller, with an initial value range of 0.3 at the lower limit and 1.5 at the upper limit. After the equivalent altitude estimate is calculated in step S3, it is first read in meters. Then, the prior value range is dynamically adjusted based on the equivalent altitude estimate: if the equivalent altitude estimate is below 2000 meters, the lower limit is adjusted to 0.4 and the upper limit to 1.2; if the equivalent altitude estimate is between 2000 and 3500 meters, the lower limit is adjusted to 0.5 and the upper limit to 1.0; if the equivalent altitude estimate is above 3500 meters, the lower limit is adjusted to 0.6 and the upper limit to 0.9. After adjustment, nine discrete candidate coefficients are selected at equal intervals between the adjusted lower and upper limits. For example, with a lower limit of 0.5 and an upper limit of 1.0, the nine candidate coefficients are 0.50, 0.56, 0.63, 0.69, 0.75, 0.81, 0.88, 0.94, and 1.00, respectively. After selection, these nine candidate coefficients are stored sequentially in a temporary array.

[0053] Substitute each candidate coefficient into the rising envelope of the current frequency to predict the change in pressure ratio at the next moment, and calculate the deviation between the predicted value and the real-time measured value of pressure ratio.

[0054] The controller pre-stores a reference frequency rising envelope, defined as a step increase every 0.5 seconds over the next 5 seconds, starting from the current frequency value, for a total of 10 steps. The frequency increments corresponding to each step are 0.2 Hz, 0.3 Hz, 0.4 Hz, 0.5 Hz, 0.6 Hz, 0.6 Hz, 0.5 Hz, 0.4 Hz, 0.3 Hz, and 0.2 Hz, respectively. For each candidate coefficient in the temporary array, the coefficient is first multiplied by the aforementioned 10 frequency increments to obtain a set of scaled step frequency increments. Then, based on the scaled step frequency increments, the voltage ratio change at the next moment (i.e., 0.5 seconds later) is predicted.

[0055] After obtaining the predicted pressure ratio change, the real-time measured pressure ratio value at the current moment is read. The predicted pressure ratio change is added to the current measured pressure ratio value to obtain the predicted pressure ratio value for the next moment. Simultaneously, the actual measured pressure ratio value at the next moment is continuously collected. The predicted pressure ratio value is subtracted from the actual measured pressure ratio value, and the absolute value is taken to obtain the change deviation corresponding to the candidate coefficient.

[0056] Repeat the above process until the variation deviation of each of the nine candidate coefficients is calculated.

[0057] Based on the magnitude of the change deviation corresponding to each candidate coefficient, a posterior weight is assigned to each candidate coefficient. The candidate coefficient with the largest posterior weight is used as the corrected heat demand tracking coefficient. Simultaneously, the prior value range for the next search is narrowed down using this heat demand tracking coefficient as the center. Specifically, for the nine candidate coefficients, the smallest change deviation value is first found and denoted as the minimum deviation. Then, for each candidate coefficient, its posterior weight is calculated: the posterior weight equals the minimum deviation divided by the change deviation of the candidate coefficient. Following this method, the smaller the change deviation of the candidate coefficient, the larger its posterior weight; the candidate coefficient whose change deviation equals the minimum deviation has a posterior weight of 1. After calculating the posterior weights of all candidate coefficients, the candidate coefficient with the largest posterior weight is found and used as the corrected heat demand tracking coefficient, and output to step S5.

[0058] Simultaneously, using this heat demand tracking coefficient as the center, a priori value range for the next search is set: the center value minus 0.1 is used as the new lower limit, and the center value plus 0.1 is used as the new upper limit. If the new lower limit is lower than 0.3, the lower limit is fixed at 0.3; if the new upper limit is higher than 1.5, the upper limit is fixed at 1.5. This new value range will be used in the first segment operation of the next round, step S4.

[0059] The current frequency rising envelope is discretized into multiple rising steps according to time, each step corresponding to a frequency increment. Specifically, the controller internally treats the frequency rise process as a time-discrete event. Starting from the current moment, with a basic time step of 0.5 seconds, a total of 10 time steps are planned, corresponding to the rising process over the next 5 seconds. Each time step is called a rising step. Within each rising step, the compressor frequency increases by a fixed increment. These 10 increments are in chronological order: first step 0.2 Hz, second step 0.3 Hz, third step 0.4 Hz, fourth step 0.5 Hz, fifth step 0.6 Hz, sixth step 0.6 Hz, seventh step 0.5 Hz, eighth step 0.4 Hz, ninth step 0.3 Hz, and tenth step 0.2 Hz. These increments are pre-stored in the controller's read-only memory, forming the reference frequency rising envelope.

[0060] For each candidate coefficient, the frequency increment of each ascending step is scaled using the candidate coefficient to obtain a scaled step sequence. This sequence is then accumulated to predict the total frequency value at the next time step. Specifically, this includes: Let the candidate coefficient being processed be... For the first A ladder of ascent ( From 1 to 10), its reference frequency increment is .Will Multiply The scaled frequency increment is obtained. Only the scaled increment of the first ascending step is taken (i.e., This is used as the frequency increment for the next time step (i.e., 0.5 seconds later). This is because the next time step refers only to a single time step after 0.5 seconds, not the entire 5 seconds. Therefore, the formula for predicting the total frequency value for the next time step is: ;in, This represents the predicted frequency value at the next moment, in Hertz (Hz). This indicates the actual operating frequency of the compressor at the current moment, in Hertz (Hz). This represents the candidate coefficient, which is dimensionless. This represents the frequency increment of the first rising step in the rising envelope of the reference frequency, which is fixed at 0.2 Hz.

[0061] Based on the differential relationship between the current real-time measured pressure ratio and the total frequency value, the pressure ratio change at the next moment is estimated using the forward differential approximation method. Specifically, the controller stores a pressure ratio-frequency differential coefficient table, which records the approximate proportional relationship between the pressure ratio change and the frequency change under different equivalent altitude estimates.

[0062] Based on the equivalent altitude estimate calculated in step S3, the corresponding differential coefficient k is obtained from the table, with the unit being the reciprocal of Hertz. Then, the predicted frequency increase value (i.e. Multiplying this by the differential coefficient, and then by a correction factor of 0.8 (to compensate for overshoot error in the forward differential), yields the predicted value of the pressure ratio change at the next moment. The calculation formula is as follows: ;in, This represents the predicted change in pressure ratio at the next moment, and is dimensionless. This represents the pressure ratio-frequency differential coefficient obtained from the equivalent altitude estimate by looking up a table, in negative first power of Hertz; Represents candidate coefficients, which are dimensionless; This represents the frequency increment of the first rising step in the rising envelope of the reference frequency, which is 0.2 Hz. For example, when the equivalent altitude estimate is 3000 meters, referring to the table, we get... The value is 0.15, if the candidate coefficient If it is 0.8, then This means the predicted pressure ratio will increase by 0.0192. This predicted value is then used to calculate the variation deviation.

[0063] Example 6: In S5, the corrected heat demand tracking coefficient is used as the load optimization benchmark for the compressor frequency upper limit and frequency ramp rate. Simultaneously, pressure ratio fluctuations are monitored in real time to iteratively update the boundary coefficients of the altitude-pressure ratio mapping relationship. Specifically, this includes: The controller continuously acquires real-time pressure ratio measurements over multiple control cycles, calculating the magnitude and direction of change between adjacent cycles. Specifically, at the end of each control cycle (cycle length 2 seconds), the controller reads the pressure values ​​measured by the compressor's discharge and suction pressure sensors, divides the discharge pressure value by the suction pressure value, and obtains the real-time pressure ratio measurement for that cycle. The controller internally includes a circular buffer with a length of 10 cycles to store the pressure ratio measurement values ​​for the most recent 10 cycles.

[0064] After obtaining a new measured pressure ratio value, subtract it from the measured pressure ratio value of the previous period. The difference is the change amplitude. If the difference is positive, the change direction is determined to be upward; if the difference is negative, the change direction is determined to be downward. At the same time, the absolute value of the change amplitude is stored in a buffer for use in subsequent steps.

[0065] When the direction of change deviates from the pressure ratio change trend predicted by the current mapping relationship, the change amplitude is multiplied by a preset attenuation factor to obtain the adjustment step size of the boundary coefficient. Specifically, the controller stores the altitude-pressure ratio mapping relationship established in step S3. This mapping relationship can predict the expected change trend (increasing or decreasing) of the pressure ratio based on the current equivalent altitude estimate. The measured change direction of the current cycle is compared with the predicted trend: if they are consistent, no adjustment is made, and the adjustment step size is set to zero; if they are opposite (i.e., the measured direction deviates from the predicted trend), the change amplitude of the current cycle is read (absolute value), multiplied by a preset attenuation factor, and the product is used as the adjustment step size of the boundary coefficient. The unit of the adjustment step size is dimensionless, and its value range is usually between 0 and 0.05.

[0066] The upper and lower boundary coefficients in the altitude-pressure ratio mapping relationship are increased or decreased in the same direction according to the adjustment step size, so that the updated boundary coefficients surround the extreme points of the current pressure ratio fluctuation. Specifically, the altitude-pressure ratio mapping relationship contains a set of upper boundary coefficients and a set of lower boundary coefficients, which correspond to the upper and lower envelopes of the pressure ratio fluctuation, respectively.

[0067] When the measured change direction is upward and deviates from the predicted trend, it indicates that the current pressure ratio is positively exceeding the upper limit of the mapping relationship. In this case, the upper limit boundary coefficient is increased by the adjustment step size, and the lower limit boundary coefficient is also increased by half of the adjustment step size. When the measured change direction is downward and deviates from the predicted trend, it indicates that the current pressure ratio is negatively exceeding the lower limit of the mapping relationship. In this case, the upper limit boundary coefficient is decreased by half of the adjustment step size, and the lower limit boundary coefficient is decreased by the adjustment step size. After the above adjustments, the interval formed by the new upper and lower limit boundary coefficients can include the pressure ratio extreme points of the current cycle and several previous cycles, thereby dynamically tracking the range of actual pressure ratio fluctuations.

[0068] The system records the cumulative number of deviations between the current pressure ratio change direction and the predicted trend. Specifically, a counter is set internally in the controller, initially set to 0. Whenever a change direction deviates from the predicted trend within a control cycle, the counter increments by 1; whenever the change direction aligns with the predicted trend, the counter is immediately reset to zero. This counter records the number of consecutive deviations since the most recent alignment.

[0069] For example, if divergence occurs in three consecutive cycles, the cumulative number of divergences is 3; if the direction is consistent in the fourth cycle, the counter is reset to zero and counting restarts. The maximum cumulative number of divergences is 10; any number exceeding 10 will be counted as 10.

[0070] The reciprocal of the cumulative number of deviations is used as the initial reference for the attenuation factor. When the cumulative number of deviations exceeds the upper limit, the initial reference is locked to the lower limit. Specifically, this involves using the reciprocal of the current cumulative number of deviations as the initial reference for the attenuation factor. For example, when the cumulative number of deviations is 1, the reciprocal is 1.0; when the cumulative number of deviations is 2, the reciprocal is 0.5; when the cumulative number of deviations is 3, the reciprocal is 0.333; and when the cumulative number of deviations is 5, the reciprocal is 0.2. Simultaneously, the lower limit of the attenuation factor is set to 0.05, and the upper limit is set to 1.0. If the cumulative number of deviations reaches the upper limit of 10, its reciprocal is 0.1, which is still greater than the lower limit of 0.05, so locking is unnecessary. If the initial reference calculated according to the reciprocal rule is lower than 0.05, then the initial reference is forcibly locked to 0.05. This step ensures that the attenuation factor is not too small, resulting in an excessively weak adjustment step.

[0071] The current change magnitude is compared with the exponentially weighted average of historical change magnitudes. Based on the comparison result, the initial benchmark is adjusted to obtain the final attenuation factor, ensuring that the larger the change magnitude, the smaller the attenuation factor. Specifically, the controller maintains an exponentially weighted average of historical change magnitudes, updated as follows: the new exponentially weighted average equals 0.9 times the exponentially weighted average of the previous period plus 0.1 times the change magnitude of the current period. The initial exponentially weighted average is set to 0.01. After obtaining the current change magnitude, it is compared with the current exponentially weighted average: if the current change magnitude is greater than the exponentially weighted average, it indicates that the current pressure ratio fluctuation is intensifying, and the initial benchmark is multiplied by 0.7 as the final attenuation factor; if the current change magnitude is less than or equal to the exponentially weighted average, it indicates that the fluctuation is stabilizing or weakening, and the initial benchmark is multiplied by 1.0 as the final attenuation factor (i.e., remains unchanged).

[0072] Through the above adjustments, when the pressure ratio changes significantly, the attenuation factor is further reduced, thereby decreasing the adjustment step size of the boundary coefficient and avoiding over-correction; when the change is small, the attenuation factor remains unchanged, ensuring the normal updating of the boundary coefficient. The final attenuation factor is the preset attenuation factor.

[0073] Example 7, please refer to Figure 2 As shown, an air source heat pump load optimization system for high-altitude environments includes: The multi-source feature acquisition module is used to collect real-time atmospheric pressure measurements, positioning altitude data, fan current change rate, and evaporator superheat dynamic response values ​​under high-altitude conditions, which are then combined into a multi-source feature sequence. The blockage confidence determination module calculates the cumulative altitude deviation based on the atmospheric pressure measurement value and the positioning altitude data in the multi-source feature sequence, and determines the blockage confidence of the measurement channel based on the cumulative altitude deviation. The equivalent altitude estimation module discards atmospheric pressure measurements when the blockage confidence exceeds a preset threshold. Instead, it substitutes the location altitude data, fan current change rate, and superheat dynamic response value into the pre-constructed altitude-pressure ratio mapping relationship to obtain the equivalent altitude estimation. The heat demand tracking coefficient correction module calculates the load matching frequency rising envelope based on the equivalent altitude estimate and the real-time measured pressure ratio, and uses Bayesian search to dynamically correct the heat demand tracking coefficient of the envelope. The load optimization benchmark control module uses the corrected heat demand tracking coefficient as the load optimization benchmark for the compressor frequency upper limit and frequency ramp rate, while monitoring the pressure ratio fluctuation in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship.

[0074] The working principle of this invention is as follows: First, atmospheric pressure measurements, location altitude data, fan current change rate, and evaporator superheat dynamic response values ​​are collected at high altitudes and combined into a multi-source feature sequence. Then, the cumulative altitude deviation is calculated based on the atmospheric pressure measurements and location altitude data in the multi-source feature sequence, and the blockage confidence of the measurement channel is determined based on this cumulative deviation. When the blockage confidence exceeds a preset threshold, the atmospheric pressure measurements are discarded, and the location altitude data, fan current change rate, and superheat dynamic response values ​​are substituted into a pre-constructed altitude-pressure ratio mapping relationship to obtain an equivalent altitude estimate. Next, based on the equivalent altitude estimate and the real-time measured pressure ratio, the rising envelope of the load matching frequency is calculated, and the heat demand tracking coefficient of the envelope is dynamically corrected using a Bayesian search. Finally, the corrected heat demand tracking coefficient is used as the load optimization benchmark for the compressor frequency upper limit and frequency rise rate, while the pressure ratio fluctuation is monitored in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship, thereby achieving load optimization of the air source heat pump in high-altitude environments.

[0075] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for optimizing the load of an air-source heat pump in a high-altitude environment, characterized in that, Includes the following steps: S1: Collect real-time atmospheric pressure measurements, location altitude data, fan current change rate, and evaporator superheat dynamic response values ​​under high-altitude environment, and combine them into a multi-source feature sequence. S2: Calculate the cumulative altitude deviation based on the atmospheric pressure measurement and positioning altitude data in the multi-source feature sequence, and determine the blockage confidence of the measurement channel based on the cumulative altitude deviation. S3: When the blockage confidence exceeds the preset threshold, the atmospheric pressure measurement value is discarded, and the positioning altitude data, the fan current change rate and the superheat dynamic response value are substituted into the pre-constructed altitude-pressure ratio mapping relationship to obtain the equivalent altitude estimate. S4: Based on the equivalent altitude estimate and the real-time measured pressure ratio, calculate the rising envelope of the load matching frequency, and use Bayesian search to dynamically correct the heat demand tracking coefficient of the envelope. S5: The corrected heat demand tracking coefficient is used as the load optimization benchmark for the compressor frequency upper limit and frequency increase rate, while the pressure ratio fluctuation is monitored in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship. S4 specifically includes: The prior range of the heat demand tracking coefficient is determined based on the equivalent altitude estimate, and multiple discrete candidate coefficients are selected evenly within the prior range. Substitute each candidate coefficient into the rising envelope of the current frequency to predict the change in pressure ratio at the next moment, and calculate the deviation between the predicted value and the real-time measured value of pressure ratio. Based on the magnitude of the change deviation corresponding to each candidate coefficient, a posterior weight is assigned to each candidate coefficient. The candidate coefficient with the largest posterior weight is used as the corrected hot demand tracking coefficient. At the same time, the prior value range of the next search is narrowed with the hot demand tracking coefficient as the center. The prediction of the pressure ratio change at the next moment specifically includes: Discretize the current frequency rising envelope into multiple rising steps according to time, with each step corresponding to a frequency increment. For each candidate coefficient, the frequency increment of each ascending step is scaled using the candidate coefficient to obtain the scaled step sequence, and the sequence is accumulated to predict the total frequency value at the next moment. Based on the differential relationship between the current real-time measured pressure ratio and the total frequency value, the pressure ratio change at the next moment is estimated using the forward difference approximation method. S5 specifically includes: The real-time measured pressure ratio values ​​were continuously collected over multiple control cycles, and the magnitude and direction of change of the measured pressure ratio values ​​between two adjacent cycles were calculated. When the direction of change deviates from the pressure ratio change trend predicted by the current mapping relationship, the change amplitude is multiplied by a preset attenuation factor to obtain the adjustment step size of the boundary coefficient. By adjusting the step size in the same direction to increase or decrease the upper and lower boundary coefficients in the altitude-pressure ratio mapping relationship, the updated boundary coefficients can surround the extreme points of the current pressure ratio fluctuation.

2. The method for optimizing the load of an air source heat pump in a high-altitude environment according to claim 1, characterized in that, S2 specifically includes: The atmospheric pressure measurements within multiple consecutive sampling periods are converted into measured altitudes, and the instantaneous difference between the measured altitude and the location altitude data in each sampling period is calculated. Each instantaneous difference is assigned a weighting coefficient that decays exponentially over time and then accumulated to obtain the cumulative dynamic altitude deviation. The cumulative dynamic altitude deviation is compared with the pre-stored unblocked deviation range, and the normalized blockage confidence score is output.

3. The method for optimizing the load of an air-source heat pump in a high-altitude environment according to claim 2, characterized in that, The normalized blocking confidence of the output specifically includes: Determine the positional offset direction of the cumulative dynamic altitude deviation relative to the lower and upper limits of the unblocked deviation range; The corresponding mapping curve is selected based on the direction of position offset. The mapping curve outputs 0 confidence when the cumulative deviation is below the lower limit and outputs single-position confidence when it is above the upper limit. When the cumulative deviation is between the lower and upper limits, the confidence level is smoothly increased from 0 to single-location confidence level according to the ratio of the distance of the cumulative deviation from the lower limit to the interval width, thus obtaining the normalized blockage confidence level.

4. The method for optimizing the load of an air source heat pump in a high-altitude environment according to claim 1, characterized in that, S3 specifically includes: Collect location altitude data, fan current change rate, superheat dynamic response value and corresponding measured pressure ratio of heat pump at different altitudes under non-clogging conditions to form multiple sets of mapping sample pairs; The ratio of the rising slope of the superheat dynamic response value to the change rate of the fan current in each mapping sample pair is extracted as an altitude-sensitive feature. The altitude-sensitive feature is then subjected to exponential fitting with the positioning altitude data to obtain the altitude-pressure ratio mapping relationship. Substitute the current location altitude data, the rate of change of wind turbine current, and the dynamic response value of superheat into the altitude-pressure ratio mapping relationship to calculate the corresponding equivalent altitude estimate.

5. The method for optimizing the load of an air source heat pump in a high-altitude environment according to claim 4, characterized in that, The step of exponentially fitting altitude-sensitive features with positioning altitude data to obtain the altitude-pressure ratio mapping relationship specifically includes: Using location elevation data as the independent variable and elevation-sensitive features as the dependent variable, an initial exponential relationship is established, which includes undetermined base parameters and proportional parameters. According to the order of the positioning altitude data from smallest to largest, each sample pair is substituted in one by one to calculate the fitting deviation under the current exponential relationship. Based on the sign of the product of the fitting deviations of two adjacent sample points, the value direction of the base parameter is adjusted in reverse. When the sum of the absolute values ​​of the fitting deviations of all sample points no longer decreases after two consecutive iterations, the current base parameter and ratio parameter are fixed, and exponential fitting is completed to obtain the altitude-pressure ratio mapping relationship.

6. The method for optimizing the load of an air-source heat pump in a high-altitude environment according to claim 5, characterized in that, The calculation process for the preset attenuation factor is as follows: Record the cumulative number of deviations between the current pressure ratio change direction and the predicted trend of the continuous deviation mapping relationship; The reciprocal of the cumulative number of divergences is used as the initial reference for the attenuation factor, and the initial reference is locked to the lower limit when the cumulative number of divergences exceeds the upper limit. The current change magnitude is compared with the exponentially weighted average of historical change magnitudes. The initial benchmark is adjusted based on the comparison results to obtain the final attenuation factor, so that the larger the change magnitude, the smaller the attenuation factor.

7. An air-source heat pump load optimization system for high-altitude environments, characterized in that, A method for optimizing the load of an air-source heat pump in a high-altitude environment as described in any one of claims 1-6, comprising: The multi-source feature acquisition module is used to collect real-time atmospheric pressure measurements, positioning altitude data, fan current change rate, and evaporator superheat dynamic response values ​​under high-altitude conditions, which are then combined into a multi-source feature sequence. The blockage confidence determination module calculates the cumulative altitude deviation based on the atmospheric pressure measurement value and the positioning altitude data in the multi-source feature sequence, and determines the blockage confidence of the measurement channel based on the cumulative altitude deviation. The equivalent altitude estimation module discards atmospheric pressure measurements when the blockage confidence exceeds a preset threshold. Instead, it substitutes the location altitude data, fan current change rate, and superheat dynamic response value into the pre-constructed altitude-pressure ratio mapping relationship to obtain the equivalent altitude estimation. The heat demand tracking coefficient correction module calculates the load matching frequency rising envelope based on the equivalent altitude estimate and the real-time measured pressure ratio, and uses Bayesian search to dynamically correct the heat demand tracking coefficient of the envelope. The load optimization benchmark control module uses the corrected heat demand tracking coefficient as the load optimization benchmark for the compressor frequency upper limit and frequency ramp rate, while monitoring the pressure ratio fluctuation in real time to iteratively update the boundary coefficient of the altitude-pressure ratio mapping relationship.

Images