A state-aware-based remote management and control system and method for industrial equipment
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SUMET INTELLIGENT TECH (JIANGSU) CO LTD
- Filing Date
- 2026-04-01
- Publication Date
- 2026-06-23
Smart Images
Figure CN122260985A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of industrial process control technology, and in particular to a remote control system and method for industrial equipment based on state awareness. Background Technology
[0002] In industrial settings, critical rotating machinery such as centrifugal pumps, compressors, and fans typically employ variable frequency drives (VFDs) for start-stop control and operational maintenance to meet the dynamic demands of production processes on flow rate, pressure, and conveying capacity. With the development of industrial digitalization and remote operation and maintenance technologies, maintenance personnel primarily collect macroscopic operational data such as electrical power, inlet and outlet pressure, and real-time flow rate through remote monitoring platforms. Based on this data, they issue speed control commands to maintain preset production capacity targets. However, during long-term operation, such equipment inevitably encounters complex physical damage processes such as cavitation, bubble collapse impact, impeller pitting, and mechanical component friction. These processes not only involve the accumulation of high-frequency vibration energy but also alter the efficiency of electrical energy conversion to hydraulic potential energy, posing potential hazards to the safe operation of the equipment.
[0003] Existing remote control methods mostly employ single-loop feedback control mechanisms based on macroscopic operating conditions. For example, they directly adjust the output frequency of the frequency converter based on the deviation between the target flow rate and the real-time flow rate to eliminate control errors. Although this control logic can ensure that process parameters follow the set target, it lacks the ability to directly perceive and constrain the internal physical damage mechanism of the equipment. When the equipment is in a hidden fault state such as the initial stage of cavitation or the aggravation of mechanical wear, the macroscopic flow or pressure indicators may not show significant abnormalities, or the automatic adjustment of the frequency converter may mask the deterioration trend of the operating condition. If the control system still forcibly increases the speed or maintains high load operation only for the purpose of maintaining the flow rate, it will often lead to further aggravation of the impact damage and energy loss inside the equipment, causing the cumulative damage of the impeller and seals to accelerate, thereby triggering more serious unplanned shutdown accidents. Summary of the Invention
[0004] The purpose of this invention is to address the shortcomings of existing technologies where remote control relies solely on macroscopic operating parameters such as flow rate and pressure for feedback, lacking effective perception and constraint of the internal physical damage mechanisms of equipment. This leads to the forced maintenance of operating parameters under hidden fault conditions, thereby accelerating the accumulation of equipment damage. Therefore, this invention proposes a state-aware remote control system and method for industrial equipment.
[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution:
[0006] A method for remote control of industrial equipment based on state awareness, comprising:
[0007] S1. Acquire the operating data of the industrial equipment in the current sampling period, the operating data including the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power and current frequency conversion frequency;
[0008] S2. Determine the high-frequency starting point based on the frequency domain distribution of the vibration acoustic sequence, and calculate the high-frequency energy ratio based on the high-frequency starting point;
[0009] S3. Calculate the hydraulic output power based on the inlet and outlet pressures and real-time flow rate, and determine the energy mismatch based on the input power and the hydraulic output power.
[0010] S4. The high-frequency energy ratio and energy mismatch are fused to obtain the state joint strength;
[0011] S5. Construct an adaptive forgetting factor based on the current frequency conversion and sampling period, and dynamically update the reference value and fluctuation scale of the joint state strength and real-time flow according to the adaptive forgetting factor.
[0012] S6. Calculate the flow frequency sensitivity and intensity frequency sensitivity based on the differential data of adjacent sampling times, and calculate the frequency update amount of the variable frequency in combination with the control weight determined by the fluctuation scale to generate remote control commands.
[0013] Preferably, determining the high-frequency starting point based on the frequency domain distribution of the vibration acoustic sequence includes:
[0014] The power spectrum sequence is obtained by performing a Fourier transform on the vibration acoustic sequence;
[0015] Calculate the logarithmic power spectrum of the power spectrum sequence, and perform second-order difference processing on the logarithmic power spectrum. The frequency corresponding to the point where the absolute value of the second-order difference processing result is the maximum is taken as the high-frequency starting point.
[0016] Calculate the total high-frequency energy from the high-frequency starting point to the highest usable frequency range of the power spectrum sequence, and the total full-frequency energy across the entire frequency range;
[0017] The proportion of high-frequency energy is obtained by the ratio of the total high-frequency energy to the total full-frequency energy.
[0018] Preferably, determining the energy mismatch based on the input power and the hydraulic output power includes:
[0019] Calculate the pressure difference between the outlet pressure and the inlet pressure, and obtain the hydraulic output power by multiplying the pressure difference by the real-time flow rate;
[0020] The difference between the input power and the hydraulic output power is calculated, and the difference is normalized based on the input power to obtain the energy mismatch.
[0021] Preferably, the high-frequency energy ratio and energy mismatch are fused, including:
[0022] The joint state strength is obtained by multiplying the high-frequency energy ratio by the energy mismatch.
[0023] Preferably, an adaptive forgetting factor is constructed, and the reference values of the joint state strength and real-time flow are dynamically updated based on the adaptive forgetting factor, including:
[0024] Multiply the current frequency by the sampling period and take the negative value. Use this negative value as the exponent of the natural constant to perform an exponentiation operation to obtain the adaptive forgetting factor.
[0025] The reference value of real-time traffic at the previous moment is weighted and summed with the real-time traffic at the current moment based on an adaptive forgetting factor to update the reference value of real-time traffic.
[0026] The reference value of the joint state strength at the previous time step is weighted and summed with the joint state strength at the current time step based on the adaptive forgetting factor, so as to update the reference value of the joint state strength.
[0027] Preferably, the fluctuation scale of the joint state strength and real-time flow is dynamically updated according to the adaptive forgetting factor, including:
[0028] Calculate the absolute value of the difference between the current state joint strength and its corresponding reference value, and recursively filter the absolute value of the difference using an adaptive forgetting factor to obtain the fluctuation scale of the state joint strength;
[0029] Calculate the absolute value of the difference between the real-time traffic at the current moment and its corresponding reference value, and recursively filter the absolute value of the difference using an adaptive forgetting factor to obtain the fluctuation scale of the real-time traffic.
[0030] Preferably, the flow sensitivity is the rate of change of the flow difference between the current time and the previous time relative to the frequency difference; the intensity sensitivity is the rate of change of the state joint intensity difference between the current time and the previous time relative to the frequency difference; the control weight includes flow control weight and risk control weight; the flow control weight is the reciprocal of the square of the fluctuation scale of the real-time flow; the risk control weight is the reciprocal of the square of the fluctuation scale of the state joint intensity.
[0031] Preferably, calculating the frequency update amount includes:
[0032] The deviation between the target flow rate and the current flow rate, as well as the deviation between the joint state strength and the reference value of the joint state strength, are used as control objectives. A squared deviation objective function weighted by flow control weight and strength control weight is constructed, and the frequency update amount is obtained by minimizing the objective function.
[0033] Preferably, generating remote control commands includes:
[0034] The frequency to be executed is obtained by adding the frequency update amount to the frequency at the current moment;
[0035] By constraining the frequency to be executed within the range of the device's rated frequency, remote control commands are obtained.
[0036] To address the above problems, the present invention also provides a state-aware remote control system for industrial equipment, the system comprising:
[0037] The data acquisition module is used to acquire the operating data of industrial equipment in the current sampling period. The operating data includes the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power, and current frequency conversion frequency.
[0038] The high-frequency extraction module is used to determine the high-frequency starting point based on the frequency domain distribution of the vibratory acoustic sequence, and to calculate the high-frequency energy ratio based on the high-frequency starting point.
[0039] The energy mismatch module is used to calculate the hydraulic output power based on the inlet and outlet pressures and real-time flow rate, and to determine the energy mismatch amount based on the input power and the hydraulic output power.
[0040] The state joint module is used to fuse the high-frequency energy ratio and energy mismatch to obtain the state joint strength;
[0041] The adaptive update module is used to construct an adaptive forgetting factor based on the current frequency conversion and sampling period, and to dynamically update the reference values and fluctuation scales of the joint state strength and real-time flow according to the adaptive forgetting factor.
[0042] The frequency conversion control decision module is used to calculate the flow frequency sensitivity and intensity frequency sensitivity based on the differential data of adjacent sampling times, and to calculate the frequency conversion update amount in combination with the control weight determined by the fluctuation scale, so as to generate remote control commands.
[0043] Compared with the prior art, the beneficial effects of the present invention are:
[0044] 1. This invention integrates the proportion of high-frequency energy extracted from the frequency domain distribution of the vibration acoustic sequence with the energy mismatch determined based on the difference between input power and hydraulic output power. This constructs a state joint strength that can simultaneously characterize the degree of enhanced impact of bubble collapse and energy conservation damage inside the equipment. It transforms the hidden microscopic physical damage mechanism into a calculable comprehensive state strength and establishes a direct mapping relationship between macroscopic operating data and microscopic damage characteristics. Thus, without adding invasive detection equipment, it achieves real-time perception and quantification of hidden fault states such as cavitation initiation, bubble collapse and mechanical wear. This overcomes the limitation that relying solely on macroscopic indicators such as flow rate or pressure can easily mask the deterioration trend of the internal health condition of the equipment.
[0045] 2. This invention utilizes an adaptive forgetting factor constructed based on the current frequency conversion and sampling period to achieve dynamic updates of the joint state strength, real-time flow reference value, and fluctuation scale. It can automatically adapt to equipment aging drift and changes in operating environment without relying on fixed alarm thresholds. Furthermore, by calculating the flow frequency sensitivity and intensity frequency sensitivity, and combining the control weights determined by the fluctuation scale to construct a weighted objective function containing flow deviation and risk state deviation, physical damage risk is directly embedded as a constraint term into the closed-loop calculation of the frequency conversion update law. This enables the remote control strategy to automatically adjust control commands according to changes in risk weights while tracking the process target flow to suppress abnormal growth in joint state strength. This effectively avoids the accelerated cumulative damage caused by forcibly maintaining production capacity under fault conditions, significantly improving the safety and reliability of remote operation of industrial equipment. Attached Figure Description
[0046] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:
[0047] Figure 1 A flowchart illustrating a state-aware remote control method for industrial equipment according to an embodiment of the present invention;
[0048] Figure 2 This is a functional block diagram of a state-aware remote control system for industrial equipment, provided as an embodiment of the present invention. Detailed Implementation
[0049] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0050] Example: This example provides a method for remote control of industrial equipment based on state awareness. See [link to example]. Figure 1 Specifically, including:
[0051] S1. Acquire the operating data of the industrial equipment in the current sampling period, the operating data including the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power and current frequency conversion frequency;
[0052] In an embodiment of the present invention, acquiring the operating data of industrial equipment in the current sampling period includes:
[0053] Acquire the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power, and current frequency of industrial equipment during the current sampling period;
[0054] It should be noted that a vibration acoustic sequence refers to a time series formed by continuously recording the mechanical wave signals generated by structural vibration and medium acoustic radiation during the operation of industrial equipment according to a sampling period. This sequence is acquired by accelerometers, acoustic emission sensors, or microphones installed on the pump casing or base. The sequence amplitude reflects the change in instantaneous vibration acceleration or sound pressure over time, and is used to characterize changes in the stress on the equipment structure and the high-frequency energy distribution changes caused by phenomena such as impact and cavitation collapse. Inlet and outlet pressures refer to the magnitude of the force per unit area exerted by the medium on the pipe wall or pressure tap on the inlet and outlet sides of the equipment. These are typically measured by inlet and outlet pressure sensors respectively, and the difference corresponds to the pressure supplied by the equipment to the medium. The pressure energy increment; real-time flow rate refers to the volume or mass flux of the medium passing through the cross-section of the equipment or pipe section per unit time, usually measured by a flow meter and updated in each sampling period, used to characterize the current conveying capacity and operating load level of the equipment; input power refers to the amount of electrical power absorbed from the power grid and transferred to the equipment by the drive motor or frequency converter, usually calculated by the voltage, current and power factor measured inside the power meter or frequency converter, reflecting the rate at which the equipment consumes energy to overcome hydraulic resistance and mechanical losses; the current frequency conversion frequency refers to the AC frequency output by the frequency converter to the motor, which directly determines the synchronous speed of the motor and thus the speed level of the equipment, used to characterize the execution status and speed control quantity of the equipment under remote control.
[0055] Specifically, a sensing network is deployed on centrifugal pumps or fans in industrial sites. The acquisition end synchronously acquires data within each current sampling period. This includes vibration acoustic sensors mounted on the outer surface of the equipment housing or bearing housing; inlet and outlet pressure sensors respectively installed at the pressure taps of the inlet and outlet pipe sections; a flow meter installed on the main pipeline; and a frequency converter communication interface or power meter interface for acquiring input power and the current frequency. The sampling period is determined by the controller based on the equipment's dynamic characteristics and signal bandwidth. Preferably, the sampling period is set to no more than one-tenth of the equipment's frequency rotation period to ensure observability of changes in operating conditions. Simultaneously, the Nyquist sampling condition corresponding to the highest frequency of interest in the vibration acoustic sequence is satisfied to avoid aliasing. The sampling frequency of the vibration acoustic sequence is constrained by the upper limit of the sensor's frequency response and the cutoff frequency of the anti-aliasing filter, and is pre-configured by the controller to be no less than twice the cutoff frequency. The acquisition end performs data acquisition at the beginning of the sampling period. Each channel is aligned to a unified timescale and triggers a multi-channel sampling task. The output of the vibration acoustic sensor is converted from analog to digital to obtain a continuous discrete vibration acoustic sequence within the sampling period. At the same timescale, the instantaneous pressure values of the inlet and outlet pressure sensors are read as the inlet and outlet pressures and range verification and zero drift compensation are performed. The instantaneous flow rate of the flow meter or the integral pulse is read to obtain the real-time flow rate and the unit conversion is consistent. The input power is directly given by the power meter or calculated from the bus voltage, bus current and power factor reported by the frequency converter. The synchronous moment value or short window average value is taken within the sampling period to suppress electrical noise. The current frequency conversion frequency is obtained from the output frequency register of the frequency converter through the industrial communication protocol and is consistent with the last transmitted frequency recorded on the controller side. Finally, the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power and current frequency conversion frequency are encapsulated into a running data frame with the same timestamp and written to the local cache and remote upload queue.
[0056] S2. Determine the high-frequency starting point based on the frequency domain distribution of the vibration acoustic sequence, and calculate the high-frequency energy ratio based on the high-frequency starting point;
[0057] In embodiments of the present invention, determining the high-frequency starting point based on the frequency domain distribution of the vibration acoustic sequence, and calculating the high-frequency energy percentage based on the high-frequency starting point, includes:
[0058] The power spectrum sequence is obtained by performing a Fourier transform on the vibration acoustic sequence;
[0059] Calculate the logarithmic power spectrum of the power spectrum sequence, and perform second-order difference processing on the logarithmic power spectrum. The frequency corresponding to the point where the absolute value of the second-order difference processing result is the maximum is taken as the high-frequency starting point.
[0060] It should be noted that the high-frequency starting point refers to the initial frequency position used to divide the high-frequency energy statistical interval after obtaining the power spectrum through frequency domain analysis of the vibration acoustic sequence. This starting frequency corresponds to the boundary point where the power spectrum transitions from low- and mid-frequency components dominated by frequency conversion and structural resonance to high-frequency components dominated by impact-type short-term events. In practice, it can be determined by finding the position with the most obvious spectral bending based on the curve shape of the logarithmic amplitude of the power spectrum as a function of frequency. The frequency corresponding to this position is determined as the high-frequency starting point, so that the frequency band above the high-frequency starting point can more concentratedly characterize the high-frequency energy contribution caused by bubble collapse impact, friction and collision, and micro-crack propagation, and can be used for subsequent calculation of the high-frequency energy ratio to reflect the degree of high-frequency impact enhancement in the equipment operation state.
[0061] Specifically, after acquiring the vibration acoustic sequence within the sampling period, the vibration acoustic sequence is recorded as a discrete sequence obtained by sampling at the sampling frequency. The sequence is first processed to remove DC bias from the sensor, and then amplitude normalization is applied to avoid interference from range differences under different operating conditions on the spectral shape judgment. Subsequently, window functions are applied to both ends of the sequence to reduce spectral leakage. The Hanning window is preferred, and its length is set to the full sequence length within the current sampling period to ensure frequency resolution and stability. Based on this, a fast Fourier transform is performed on the windowed discrete sequence to obtain a complex spectrum sequence, and the square of the amplitude of the complex spectrum sequence is taken to obtain the power spectrum sequence. Simultaneously, only the single-sided frequency band corresponding to zero frequency to half the sampling frequency is retained in the power spectrum sequence for subsequent analysis. Then, the natural logarithm is taken point-by-point on the power spectrum sequence to obtain the logarithmic power spectrum. To avoid numerical divergence when the power approaches zero during logarithmic operations, a minimum positive number is superimposed on the power spectrum sequence before taking the logarithm, and this minimum positive number is preferred. The median of the power spectrum sequence is selected as one millionth to ensure that the relative structure of the spectrum shape is not changed. Then, the logarithmic power spectrum is subjected to second-order difference processing according to the frequency index. The second-order difference can be approximated by the discrete second derivative of three adjacent points. That is, for each frequency index, the next index value is calculated, the previous index value is added to the next index value, and then twice the current index value is subtracted to obtain the second-order difference sequence characterizing the degree of spectral bending. Then, the absolute value of the second-order difference sequence is taken point by point to capture the significant changes of upward bending and downward bending at the same time. Finally, the index with the largest absolute value is found within the predefined effective search frequency band, and the frequency corresponding to the index is determined as the high-frequency starting point. The lower limit of the effective search frequency band is preferably three times the equipment switching frequency to avoid the dominant influence of the switching frequency and its low-order harmonics on the identification of bending points. The upper limit is preferably the highest available frequency of a single-sided frequency band to cover possible impact high-frequency components, so that the determined high-frequency starting point can stably fall at the transition position between the low-mid frequency dominance region and the high-frequency impact dominance region.
[0062] Calculate the total high-frequency energy from the high-frequency starting point to the highest usable frequency range of the power spectrum sequence, and the total full-frequency energy across the entire frequency range;
[0063] The proportion of high-frequency energy is obtained by the ratio of the total high-frequency energy to the total full-frequency energy.
[0064] It should be noted that the high-frequency energy ratio refers to the normalized quantity calculated by comparing the sum of the power spectrum energy from the high-frequency starting point to the highest usable frequency range with the sum of the power spectrum energy across the entire frequency range after obtaining the power spectrum sequence through frequency domain analysis of the vibration acoustic sequence. This ratio is used to characterize the relative contribution of high-frequency impact components to the overall vibration acoustic energy during equipment operation. The high-frequency energy sum is obtained by accumulating the power spectrum values at each frequency point within the high-frequency range, while the full-frequency energy sum is obtained by accumulating the power spectrum values at each frequency point across the entire frequency range. An increase in the high-frequency energy ratio indicates a relative enhancement of the high-frequency energy components caused by short-term pulse events such as local impacts, friction, and bubble collapse of the pump casing. Therefore, it can serve as an important state quantity reflecting changes in equipment status and abnormal signs.
[0065] Specifically, after obtaining the power spectrum sequence and determining the high-frequency starting point, the power spectrum sequence is limited to a single-sided frequency band consisting of zero frequency to the highest usable frequency. The highest usable frequency is determined by the sampling frequency of the vibroacoustic sequence and is preferably half of the sampling frequency to satisfy the Nyquist sampling condition and coincide with the effective passband of the front-end anti-aliasing filter. Subsequently, the frequency resolution of the power spectrum sequence is obtained, and a mapping relationship from frequency index to physical frequency is established accordingly. The frequency resolution is determined by dividing the sampling frequency by the number of points in the Fast Fourier Transform. Then, the starting index in the power spectrum sequence is obtained based on the frequency corresponding to the high-frequency starting point, and the mapping relationship from the starting index to the highest usable frequency is established. The power spectrum values of each frequency point within the corresponding end index range are accumulated point by point to obtain the high-frequency energy sum. At the same time, the power spectrum values of each frequency point from zero frequency to the end index range are accumulated point by point to obtain the full-frequency energy sum. In order to avoid the inconsistency of energy scale caused by different sampling lengths, it is preferable to normalize the above accumulation results according to the number of fast Fourier transform points to decouple the energy sum from the sampling window length. The high-frequency energy ratio is obtained by dividing the high-frequency energy sum by the full-frequency energy sum, so that the high-frequency energy ratio can stably reflect the proportion of high-frequency impact components relative to the overall vibration acoustic energy and serve as the input for subsequent state joint intensity calculation.
[0066] S3. Calculate the hydraulic output power based on the inlet and outlet pressures and real-time flow rate, and determine the energy mismatch based on the input power and the hydraulic output power.
[0067] In embodiments of the present invention, calculating the hydraulic output power and determining the energy mismatch based on the input power and the hydraulic output power includes:
[0068] Calculate the pressure difference between the outlet pressure and the inlet pressure, and obtain the hydraulic output power by multiplying the pressure difference by the real-time flow rate;
[0069] Calculate the difference between the input power and the hydraulic output power, and normalize the difference based on the input power to obtain the energy mismatch.
[0070] It should be noted that energy mismatch is a normalized quantity used to characterize the degree of inconsistency between the input energy and the effective hydraulic energy output to the medium of industrial equipment during the current sampling period. The input energy corresponds to the rate at which electrical energy is converted into mechanical energy and supplied to the equipment, as represented by the input power. The effective hydraulic energy output to the medium corresponds to the hydraulic output power, which is obtained by multiplying the real-time flow rate by the pressure difference between the inlet and outlet pressures. The energy mismatch is obtained by calculating the difference between the input power and the hydraulic output power and normalizing it to the input power. It is used to reflect the relative proportion of energy consumption other than hydraulic output. This relative proportion includes additional losses caused by mechanical friction, eddy current and leakage losses, as well as bubble formation and collapse. When the energy mismatch increases, it indicates that the unit hydraulic output decreases or the additional losses increase under the same input power. Therefore, it can be used to indicate abnormal operating conditions or enhanced internal damage mechanisms of the equipment.
[0071] Specifically, after simultaneously acquiring the inlet and outlet pressures within the current sampling period, both are standardized to the same pressure unit to avoid dimensional inconsistencies. Simultaneously, the zero point and range of the pressure sensor are validated to eliminate over-range or distorted readings. After successful validation, the pressure difference is obtained by subtracting the inlet pressure from the outlet pressure and is considered as the incremental pressure energy supplied by the equipment to the medium. Subsequently, the real-time flow rate within the same sampling period is read, and the volumetric flow rate and mass flow rate are standardized. Volumetric flow rate is preferred for calculation to align with the commonly used expression for hydraulic power derived from pressure difference. Based on this, the pressure difference is multiplied by the real-time flow rate to obtain the hydraulic output power, which is then used as the effective power output on the medium side. Next, the input power within the same sampling period is read and data integrity is checked to ensure that it is the effective power provided by the drive motor or frequency converter to the equipment. Then, the difference between the input power and the hydraulic output power is calculated to characterize the part of the input power that is not converted into hydraulic output. This difference is divided by the input power to obtain the normalized difference. Finally, the normalized difference is output as the energy mismatch, so that it can characterize the relative degree of additional loss and efficiency reduction within the current sampling period.
[0072] S4. The high-frequency energy ratio and energy mismatch are fused to obtain the state joint strength;
[0073] In embodiments of the present invention, the fusion of high-frequency energy ratio and energy mismatch includes:
[0074] The joint state strength is obtained by multiplying the high-frequency energy ratio by the energy mismatch.
[0075] Specifically, the state joint strength refers to the comprehensive state quantity obtained by fusing the high-frequency energy proportion and the energy mismatch. It is used to simultaneously characterize the relative enhancement of high-frequency impact energy components and the degree of mismatch between input energy and hydraulic output in industrial equipment within the current sampling period. The high-frequency energy proportion reflects the proportion of high-frequency energy caused by short-time pulse events such as bubble collapse impact and friction impact in the overall energy in the vibroacoustic sequence. The energy mismatch reflects the relative proportion of input power that has not been converted into hydraulic output power and indicates the degree of additional losses and efficiency reduction. The state joint strength is obtained by calculating the product of the high-frequency energy proportion and the energy mismatch. This comprehensive state quantity only increases significantly when the high-frequency impact enhancement and energy mismatch increase simultaneously. Thus, it can more focused on the abnormal state consistent with the internal impact mechanism and energy conservation violation and serve as a key state constraint quantity for remote control.
[0076] S5. Construct an adaptive forgetting factor based on the current frequency conversion and sampling period, and dynamically update the reference value and fluctuation scale of the joint state strength and real-time flow according to the adaptive forgetting factor.
[0077] In embodiments of the present invention, an adaptive forgetting factor is constructed, and the reference values and fluctuation scales of the joint state strength and real-time flow are dynamically updated based on the adaptive forgetting factor, including:
[0078] Multiply the current frequency by the sampling period and take the negative value. Use this negative value as the exponent of the natural constant to perform an exponentiation operation to obtain the adaptive forgetting factor.
[0079] It should be noted that the adaptive forgetting factor is a coefficient used to assign weights to historical and current data when dynamically updating reference values and fluctuation scales. It adapts to the current frequency conversion frequency and sampling period of industrial equipment to reflect the speed of equipment operation status updates and make the statistics follow changes in operating conditions.
[0080] Specifically, the current frequency of the inverter is read and converted to Hertz for calculation. Simultaneously, the sampling period length is obtained from the remote control controller and converted to seconds for calculation. Then, the current frequency of the inverter is multiplied by the sampling period to obtain a dimensionless quantity, which is then negative. This negative value is used as the exponent of the natural constant for power operation to obtain the adaptive forgetting factor. The formula for calculating the adaptive forgetting factor is as follows:
[0081] In the formula, This is an adaptive forgetting factor, with a value ranging from zero to one. The current frequency of the inverter. For the sampling period, the This represents the number of electrical frequency cycles within a sampling period. From the perspective of energy input and operating condition changes, a higher frequency or a longer sampling period means that the equipment experiences more excitation cycles within that period. State statistics need to reflect new data more quickly to avoid historical data masking current changes. Therefore, an exponential decay form is used to make the forgetting factor increase with... The forgetting factor increases automatically and decreases automatically, thereby reducing the weight of historical data and increasing the weight of current data. Conversely, when the frequency is low or the sampling period is short, the forgetting factor is closer to one to enhance smoothness and suppress measurement noise. Finally, the calculated adaptive forgetting factor is written into the running data frame as a subsequent reference value and a weight coefficient for recursively updated fluctuation scale.
[0082] The reference value of real-time traffic at the previous moment is weighted and summed with the real-time traffic at the current moment based on an adaptive forgetting factor to update the reference value of real-time traffic.
[0083] The reference value of the joint state strength at the previous time step is weighted and summed with the joint state strength at the current time step based on the adaptive forgetting factor to update the reference value of the joint state strength.
[0084] Specifically, after calculating the adaptive forgetting factor and obtaining the current real-time traffic and state joint strength, the previous real-time traffic reference value and the previous state joint strength reference value are read as historical statistical benchmarks. The current real-time traffic is directly assigned as the real-time traffic reference value, and the current state joint strength is directly assigned as the state joint strength reference value to complete the initial alignment. Subsequently, in the regular update process, the adaptive forgetting factor is used as the historical weight, and one minus the adaptive forgetting factor is used as the current weight. The previous real-time traffic reference value and the current real-time traffic are weighted and summed to obtain the updated real-time traffic reference value. At the same time, the previous state joint strength reference value and the current state joint strength are weighted and summed to obtain the updated state joint strength reference value. The update calculation formulas for the traffic reference value and the state joint strength reference value are as follows:
[0085] In the formula, This is the real-time traffic reference value for the current moment. This is the real-time traffic reference value from the previous moment. This represents the real-time traffic at the current moment. This is the current state joint strength reference value. The joint strength reference value of the state at the previous moment. The joint strength of the state at the current moment, As an adaptive forgetting factor, this weighted update makes the reference value numerically equivalent to an exponentially weighted moving average of historical data, thereby achieving continuous tracking of operating condition drift and suppressing the disturbance of instantaneous measurement noise to the reference benchmark. Finally, the updated real-time flow reference value and the joint state intensity reference value are written back to the state cache and stored in association with the current sampling period timestamp for subsequent fluctuation scale calculation and control weight calculation.
[0086] Calculate the absolute value of the difference between the current state joint strength and its corresponding reference value, and recursively filter the absolute value of the difference using an adaptive forgetting factor to obtain the fluctuation scale of the state joint strength;
[0087] Calculate the absolute value of the difference between the real-time traffic at the current moment and its corresponding reference value, and recursively filter the absolute value of the difference using an adaptive forgetting factor to obtain the fluctuation scale of the real-time traffic.
[0088] It should be noted that the fluctuation scale of the joint state strength refers to a statistical measure used to measure the typical deviation of the joint state strength from its reference value during continuous sampling. It is obtained by taking the absolute value of the difference between the current joint state strength and the corresponding reference value and recursively filtering it with an adaptive forgetting factor. This numerically reflects the strength of the fluctuation of the joint state strength in recent operation and suppresses the accidental influence of a single anomaly. The larger the fluctuation scale, the more obvious the fluctuation of the joint state strength around the reference value and the stronger the combined instability of abnormal shocks and energy mismatch. The fluctuation scale of real-time flow refers to a statistical measure used to measure the typical deviation of real-time flow from its real-time flow reference value during continuous sampling. It is also based on the absolute value of the difference between the current real-time flow and the corresponding reference value and recursively filtered by an adaptive forgetting factor. The larger the fluctuation scale, the more obvious the fluctuation of the flow around the reference level and the more drastic the changes in operating load or pipeline conditions. The above two types of fluctuation scales serve as the basis for subsequent control weights. When a certain quantity itself fluctuates greatly and is unstable, its weight is automatically reduced to avoid the control being pulled by noise or severe disturbances. When a certain quantity fluctuates little and is stable, its weight is automatically increased to enhance the credible constraint of the control on that quantity.
[0089] Specifically, the system reads the current state joint strength, the current real-time traffic, the current state joint strength reference value, and the current real-time traffic reference value. Simultaneously, it reads the adaptive forgetting factor calculated within the same sampling period and verifies its value range to ensure it is between zero and one. Then, it calculates the deviation of the state joint strength from its reference value and the deviation of the real-time traffic from its reference value. The deviation is expressed as the absolute value of the difference to simultaneously represent the magnitude of positive and negative deviations and avoid underestimation of fluctuations caused by sign cancellation. Upon device initial startup or when cache is missing, the absolute value of the difference is directly used as the initial value of the corresponding fluctuation scale. To ensure the availability of the recursive quantity, under normal operating conditions, the adaptive forgetting factor is used as the historical weight, and one minus the adaptive forgetting factor is used as the current weight. Recursive filtering is performed on the absolute value of the difference to obtain the fluctuation scale, making the fluctuation scale numerically equivalent to an exponentially weighted moving average of the deviation amplitude. This allows for both suppression of short-term disturbances and tracking of operating condition drift without setting a fixed window length. The recursive filtering can be implemented as follows: first, the absolute value of the difference in the joint state strength is updated to obtain the fluctuation scale of the joint state strength; then, the absolute value of the difference in the real-time flow is updated to obtain the fluctuation scale of the real-time flow. The calculation formula is as follows:
[0090] In the formula, The current state is the joint intensity fluctuation scale. The joint intensity fluctuation scale of the state at the previous moment. The joint strength of the state at the current moment, This is the current state joint strength reference value. This is a real-time traffic fluctuation scale at the current moment. This represents the real-time traffic fluctuation scale from the previous moment. This represents the real-time traffic at the current moment. This is the real-time traffic reference value for the current moment. As an adaptive forgetting factor, the above update enables the fluctuation scale to converge smoothly and adaptively with the current operating rhythm of the equipment and reflect the degree of measurement instability. Finally, the obtained state joint intensity fluctuation scale and real-time flow fluctuation scale are written into the state cache and stored in association with the current sampling period timestamp for subsequent control weight calculation and frequency conversion update calculation.
[0091] S6. Calculate the flow frequency sensitivity and intensity frequency sensitivity based on the differential data of adjacent sampling times, and calculate the frequency update amount of the variable frequency in combination with the control weight determined by the fluctuation scale to generate remote control commands.
[0092] In embodiments of the present invention, the flow frequency sensitivity and intensity frequency sensitivity are calculated, and the frequency update amount is calculated by combining the control weights determined by the fluctuation scale to generate remote control commands, including:
[0093] Calculate the flow frequency sensitivity and intensity frequency sensitivity based on the differential data from adjacent sampling times;
[0094] Flow control weights and risk control weights are calculated based on fluctuation scales;
[0095] It should be noted that flow frequency sensitivity refers to the local rate of change of the real-time flow rate corresponding to a unit change in the frequency converter. It is obtained by the ratio of the difference in real-time flow rate at adjacent sampling times to the difference in frequency converter at adjacent sampling times, reflecting the immediate impact of speed regulation on the transport capacity under the current operating conditions. Intensity frequency sensitivity refers to the local rate of change of the joint intensity of the state corresponding to a unit change in the frequency converter. It is obtained by the ratio of the difference in joint intensity of the state at adjacent sampling times to the difference in frequency converter at adjacent sampling times, reflecting the immediate impact of speed regulation on the combined state of high-frequency impact and energy mismatch under the current operating conditions. Flow control weight refers to the confidence weight assigned to the flow control target when solving for the frequency converter update, which is based on the real-time flow rate. The fluctuation scale is determined by taking the inverse of the square of the real-time flow fluctuation scale, so that when the flow measurement is stable and the fluctuation scale is small, the weight increases, thereby strengthening the constraint on the target flow. The risk control weight refers to the confidence weight assigned to the risk control target when solving the frequency update quantity. It is determined based on the fluctuation scale of the joint state strength and takes the inverse of the square of the joint state strength fluctuation scale, so that when the joint state strength measurement is stable and the fluctuation scale is small, the weight increases, thereby strengthening the constraint on the risk state. The above two types of sensitivity are used to map the change of control quantity to the change of state quantity to form a solvable update relationship. The two types of control weights are used to transform the reliability difference of different state quantities into the influence distribution in the solution process, so that remote control remains robust under the changes in operating conditions and noise disturbances.
[0096] Specifically, the system reads the real-time flow rate, the joint state strength, and the frequency conversion rate between the current and previous times. It verifies the continuity of timestamps between adjacent times to ensure the difference corresponds to the same sampling period step. Simultaneously, it verifies the validity of the frequency conversion rate difference to avoid instability caused by a constant frequency, where the denominator is zero. When the frequency conversion rate difference is detected to be zero or close to zero, it preferentially uses the sensitivity calculated from the previous valid sample pair as the current sensitivity to maintain the continuity of the control law. Then, it calculates the real-time flow rate difference between the current and previous times and divides it by the corresponding frequency conversion rate difference to obtain the flow rate sensitivity, which characterizes the local rate of change of the flow rate response to frequency conversion rate changes. Simultaneously, it calculates the joint state strength difference between the current and previous times. The intensity frequency sensitivity is obtained by dividing the value by the corresponding frequency difference, which characterizes the local rate of change of the frequency change on the joint state intensity. Then, after obtaining the fluctuation scale of the real-time flow and the fluctuation scale of the joint state intensity, the square of the fluctuation scale is used as the quantitative basis of the uncertainty intensity, and its reciprocal is used as the control weight to achieve priority of stable measurement and suppression of unstable measurement. The flow control weight is determined by the reciprocal of the square of the fluctuation scale of the real-time flow, and the risk control weight is determined by the reciprocal of the square of the fluctuation scale of the joint state intensity, so that the risk control weight automatically increases when the fluctuation of the joint state intensity is small to strengthen the constraint on the risk state. Finally, the flow frequency sensitivity, intensity frequency sensitivity, flow control weight and risk control weight are written into the running data frame as input for solving the frequency change update.
[0097] The deviation between the target flow and the current flow, and the deviation between the joint state strength and the joint state strength reference value are used as control targets. A squared deviation objective function weighted by flow control weight and strength control weight is constructed, and the frequency update amount is obtained by minimizing the objective function.
[0098] The frequency update amount refers to the incremental value calculated by the remote control side within the current sampling period to correct the output frequency of the frequency converter in the next sampling period. It uses the deviation between the target flow and the current flow, as well as the deviation between the joint state strength and its reference value, as control targets. It combines flow frequency sensitivity and intensity frequency sensitivity to perform local linear characterization of the impact of frequency changes on flow and joint state strength. Then, it combines flow control weight and risk control weight to weightedly allocate the credibility and constraint strength of the two types of control targets. Thus, a frequency correction range that takes into account both flow tracking and risk constraints is obtained by solving the problem. A positive frequency update amount indicates that the output frequency is increased to enhance the transmission capacity, while a negative frequency update amount indicates that the output frequency is decreased to weaken the excitation on the equipment and promote the return of the joint state strength to the reference level.
[0099] Specifically, the target flow rate is obtained and the flow deviation is calculated by comparing it with the real-time flow rate at the current moment. Simultaneously, the current state joint strength reference value is obtained and the strength deviation is calculated by comparing it with the current state joint strength. The target flow rate is given by the upper-level production process or set by the operator on the remote control interface. Its value is determined based on meeting the minimum supply of downstream processes and the allowable operating range of the equipment, and is constrained by the rated frequency boundary of the equipment. Then, the frequency update is used as the variable to be solved, and local linearization is used to express the impact of the frequency update on the real-time flow rate and state joint strength as an incremental relationship characterized by sensitivity. That is, the frequency update multiplied by the flow frequency sensitivity is used as the predicted change in flow rate, and the frequency update multiplied by the strength frequency sensitivity is used as the predicted change in state joint strength. Based on this, the importance and reliability of the flow deviation and strength deviation are weighted by flow control weight and risk control weight, respectively. A weighted squared deviation objective function is constructed, and minimizing this objective function yields a unique frequency update. The objective function is:
[0100]
[0101] In the formula, Let be the objective function to be minimized. This is the frequency update amount for the variable frequency drive. For target traffic, This represents the real-time traffic at the current moment. The joint strength of the state at the current moment, This is the current state joint strength reference value. Flow frequency sensitivity represents the rate of change in flow rate caused by a unit change in the frequency converter. Intensity-frequency sensitivity represents the rate of change of the combined intensity of the state caused by a unit change in the frequency of the frequency converter. Weighting is used to control traffic. The reason for using the weighted squared deviation form for risk control weights is to make the two types of control objectives comparable and analytically solvable within the same scalar objective function. Furthermore, the control weights automatically suppress the influence of unstable measurements with large fluctuations on the solution results, thereby improving the robustness of remote control under noise and operating condition disturbances. Subsequently, differentiating the objective function with respect to the frequency update and setting the derivative to zero yields a closed-form solution to avoid uncertainties introduced by iterative solutions. The closed-form solution is:
[0102]
[0103] In the formula, This is the frequency update amount for the variable frequency drive. For target traffic, This represents the real-time traffic at the current moment. The joint strength of the state at the current moment, This is the current state joint strength reference value. Flow frequency sensitivity represents the rate of change in flow rate caused by a unit change in the frequency converter. Intensity-frequency sensitivity represents the rate of change of the combined intensity of the state caused by a unit change in the frequency of the frequency converter. Weighting is used to control traffic. For risk control weights.
[0104] The frequency to be executed is obtained by adding the frequency update amount to the frequency at the current moment;
[0105] The frequency to be executed is constrained within the range of the device's rated frequency to obtain remote control commands;
[0106] Specifically, after solving for the frequency update, the frequency update is read, along with the frequency collected within the same sampling period at the current moment. Then, the frequency update and the frequency at the current moment are algebraically added to obtain the frequency to be executed. This ensures that the frequency to be executed simultaneously inherits the current execution state and is superimposed with the control correction amplitude of the current period. The addition relationship can be achieved using a conventional superposition method for frequency correction. Next, the rated frequency boundaries of the equipment are determined based on the nameplate parameters of the motor and frequency converter, as well as the on-site commissioning records. These rated frequency boundaries include a minimum rated frequency and a maximum rated frequency, where the maximum rated frequency is preferably the motor's rated frequency or the frequency converter's maximum allowable output frequency. The lower of the frequencies is selected to avoid abnormal mechanical stress and electromagnetic loss caused by overspeed operation. The minimum rated frequency is preferably selected as the minimum allowable frequency required to ensure continuous and stable delivery of pump equipment, based on the premise that no significant instability or persistent inefficiency occurs on site. Then, the frequency to be executed is bounded to the minimum rated frequency and the maximum rated frequency. When the frequency to be executed is less than the minimum rated frequency, it is set to the minimum rated frequency. When the frequency to be executed is greater than the maximum rated frequency, it is set to the maximum rated frequency, thereby obtaining the final remote control command frequency value. Finally, the remote control command is sent to the frequency converter through the industrial communication protocol and written into the audit log to achieve traceable remote control.
[0107] like Figure 2 The diagram shown is a functional block diagram of a state-aware remote control system for industrial equipment provided in an embodiment of the present invention.
[0108] In this embodiment, the functions of each module / unit are as follows:
[0109] The data acquisition module is used to acquire the operating data of industrial equipment in the current sampling period. The operating data includes the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power, and current frequency conversion frequency.
[0110] The high-frequency extraction module is used to determine the high-frequency starting point based on the frequency domain distribution of the vibratory acoustic sequence, and to calculate the high-frequency energy ratio based on the high-frequency starting point.
[0111] The energy mismatch module is used to calculate the hydraulic output power based on the inlet and outlet pressures and real-time flow rate, and to determine the energy mismatch amount based on the input power and the hydraulic output power.
[0112] The state joint module is used to fuse the high-frequency energy ratio and energy mismatch to obtain the state joint strength;
[0113] The adaptive update module is used to construct an adaptive forgetting factor based on the current frequency conversion and sampling period, and to dynamically update the reference values and fluctuation scales of the joint state strength and real-time flow according to the adaptive forgetting factor.
[0114] The frequency conversion control decision module is used to calculate the flow frequency sensitivity and intensity frequency sensitivity based on the differential data of adjacent sampling times, and to calculate the frequency conversion update amount in combination with the control weight determined by the fluctuation scale, so as to generate remote control commands.
[0115] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A method for remote control of industrial equipment based on state awareness, characterized in that, include: S1. Acquire the operating data of the industrial equipment in the current sampling period, the operating data including the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power and current frequency conversion frequency; S2. Determine the high-frequency starting point based on the frequency domain distribution of the vibration acoustic sequence, and calculate the high-frequency energy ratio based on the high-frequency starting point; S3. Calculate the hydraulic output power based on the inlet and outlet pressures and real-time flow rate, and determine the energy mismatch based on the input power and the hydraulic output power. S4. The high-frequency energy ratio and energy mismatch are fused to obtain the state joint strength; S5. Construct an adaptive forgetting factor based on the current frequency conversion and sampling period, and dynamically update the reference value and fluctuation scale of the joint state strength and real-time flow according to the adaptive forgetting factor. S6. Calculate the flow frequency sensitivity and intensity frequency sensitivity based on the differential data of adjacent sampling times, and calculate the frequency update amount of the variable frequency in combination with the control weight determined by the fluctuation scale to generate remote control commands.
2. The method for remote control of industrial equipment based on state awareness according to claim 1, characterized in that, Determining the high-frequency starting point based on the frequency domain distribution of the vibratory acoustic sequence includes: The power spectrum sequence is obtained by performing a Fourier transform on the vibration acoustic sequence; Calculate the logarithmic power spectrum of the power spectrum sequence, and perform second-order difference processing on the logarithmic power spectrum. The frequency corresponding to the point where the absolute value of the second-order difference processing result is the maximum is taken as the high-frequency starting point. Calculate the total high-frequency energy from the high-frequency starting point to the highest usable frequency range of the power spectrum sequence, and the total full-frequency energy across the entire frequency range; The proportion of high-frequency energy is obtained by the ratio of the total high-frequency energy to the total full-frequency energy.
3. The method for remote control of industrial equipment based on state awareness according to claim 1, characterized in that, The energy mismatch is determined based on the input power and the hydraulic output power, including: Calculate the pressure difference between the outlet pressure and the inlet pressure, and obtain the hydraulic output power by multiplying the pressure difference by the real-time flow rate; The difference between the input power and the hydraulic output power is calculated, and the difference is normalized based on the input power to obtain the energy mismatch.
4. The method for remote control of industrial equipment based on state awareness according to claim 1, characterized in that, The high-frequency energy ratio and energy mismatch are integrated, including: The joint state strength is obtained by multiplying the high-frequency energy ratio by the energy mismatch.
5. The method for remote control of industrial equipment based on state awareness according to claim 1, characterized in that, An adaptive forgetting factor is constructed, and the reference values of the joint state strength and real-time flow are dynamically updated based on the adaptive forgetting factor, including: Multiply the current frequency by the sampling period and take the negative value. Use this negative value as the exponent of the natural constant to perform an exponentiation operation to obtain the adaptive forgetting factor. The reference value of real-time traffic at the previous moment is weighted and summed with the real-time traffic at the current moment based on an adaptive forgetting factor to update the reference value of real-time traffic. The reference value of the joint state strength at the previous time step is weighted and summed with the joint state strength at the current time step based on the adaptive forgetting factor, so as to update the reference value of the joint state strength.
6. The method for remote control of industrial equipment based on state awareness according to claim 1, characterized in that, The joint strength of states and the fluctuation scale of real-time flow are dynamically updated based on an adaptive forgetting factor, including: Calculate the absolute value of the difference between the current state joint strength and its corresponding reference value, and recursively filter the absolute value of the difference using an adaptive forgetting factor to obtain the fluctuation scale of the state joint strength; Calculate the absolute value of the difference between the real-time traffic at the current moment and its corresponding reference value, and recursively filter the absolute value of the difference using an adaptive forgetting factor to obtain the fluctuation scale of the real-time traffic.
7. The method for remote control of industrial equipment based on state awareness according to claim 1, characterized in that, The flow sensitivity is the rate of change of the flow difference between the current time and the previous time relative to the frequency difference. The intensity sensitivity is the rate of change of the combined intensity difference between the current time and the previous time relative to the frequency difference. The control weights include flow control weights and risk control weights; The flow control weight is the reciprocal of the square of the fluctuation scale of the real-time flow; the risk control weight is the reciprocal of the square of the fluctuation scale of the joint state strength.
8. The method for remote control of industrial equipment based on state awareness according to claim 7, characterized in that, Calculate the frequency update amount for the inverter, including: The deviation between the target flow rate and the current flow rate, as well as the deviation between the joint state strength and the reference value of the joint state strength, are used as control objectives. A squared deviation objective function weighted by flow control weight and strength control weight is constructed, and the frequency update amount is obtained by minimizing the objective function.
9. The method for remote control of industrial equipment based on state awareness according to claim 1, characterized in that, Generate remote control commands, including: The frequency to be executed is obtained by adding the frequency update amount to the frequency at the current moment; By constraining the frequency to be executed within the range of the device's rated frequency, remote control commands are obtained.
10. A state-aware remote control system for industrial equipment, characterized in that, The system includes: The data acquisition module is used to acquire the operating data of industrial equipment in the current sampling period. The operating data includes the vibration acoustic sequence, inlet and outlet pressure, real-time flow rate, input power, and current frequency conversion frequency. The high-frequency extraction module is used to determine the high-frequency starting point based on the frequency domain distribution of the vibratory acoustic sequence, and to calculate the high-frequency energy ratio based on the high-frequency starting point. The energy mismatch module is used to calculate the hydraulic output power based on the inlet and outlet pressures and real-time flow rate, and to determine the energy mismatch amount based on the input power and the hydraulic output power. The state joint module is used to fuse the high-frequency energy ratio and energy mismatch to obtain the state joint strength; The adaptive update module is used to construct an adaptive forgetting factor based on the current frequency conversion and sampling period, and to dynamically update the reference values and fluctuation scales of the joint state strength and real-time flow according to the adaptive forgetting factor. The frequency conversion control decision module is used to calculate the flow frequency sensitivity and intensity frequency sensitivity based on the differential data of adjacent sampling times, and to calculate the frequency conversion update amount in combination with the control weight determined by the fluctuation scale, so as to generate remote control commands.