WO2021251809A1

WO2021251809A1 - Dynamic weighing system operating at high speed designed to measure the static load of the axles and the overall weight of a road transportation vehicle

Info

Publication number: WO2021251809A1
Application number: PCT/MA2021/000009
Authority: WO
Inventors: Lhoussaine OUBRICH; Mohammed OUASSAID; Mohamed MAAROUFI
Original assignee: Oubrich Lhoussaine
Priority date: 2020-06-12
Filing date: 2021-06-11
Publication date: 2021-12-16
Also published as: MA50051A1; WO2021251809A4; MA50051B1

Abstract

The invention relates to a dynamic weighing system operating at high speed that can measure, under normal traffic conditions, the static load of the axles and the overall weight of the road transportation vehicle (5). This system is made up of a grid of piezoelectric sensors (1), with an optimal number of ten sensors, buried transversely in the traffic lane of the roadway, where the distance between the sensors is non-uniform and defined according to the Tchebychev formula. This grid of piezoelectric sensors (1) makes it possible to reconstruct the analog signal of the load of the axle of the vehicle (5) on the basis of the voltages generated by the sensors, the values of which are proportional to the immediate forces applied by the wheels of the axle of the vehicle to the surface of the road (6). This analog signal comprises a continuous component (static load of the axle) and a variable component (dynamic loads generated by the wheel-road interaction). The sensor grid (1) is coupled with an analog filter comprising two circuit boards. The first circuit board, which performs conditioning of the signal produced by the piezoelectric sensor, processes the signal from the sensors, which is characterized by a signal-to-noise ratio that is too low, in order to increase the readability and usability of said signal. The second circuit board, which is dedicated to processing the analog signal and measuring the static load of the axle of the vehicle (5), removes, by means of a high-pass filter, the continuous component of the analog signal reconstructed from the static load of the axle and then defines the points of the filtered signal that have a null variable component. At these points, the value of the voltage measured on the reconstructed analog signal of the axle is only the voltage of the static load of the axle. By introducing the calibrating coefficient, the value of the measured voltage relative to the load of the axle is transformed into a value in weight units.

Description

HIGH-SPEED DYNAMIC WEIGHING SYSTEM

INTENDED FOR MEASURING STATIC AXLE LOAD AND OVERALL WEIGHT OF ROAD TRANSPORT VEHICLES

Description

I. Technical field

The present invention relates to the field of dynamic in-motion weighing systems at high speed intended for the control and automated sanctioning of transport trucks, in this case means of transporting goods.

II. Objective of the invention

In order to combat the phenomenon of technical overloads practiced by freight transport trucks, the invention aims to lead to a technical solution contributing to setting up a credible and reliable dynamic weighing system in operation at high speed for automated sanctioning. offenders. The envisaged solution must be operational 24 hours a day, 7 days a week for road traffic running at normal traffic speeds.

III. Scope of the invention

The invention relates to the technique of measuring the axle load of moving vehicles at high speed by introducing an intelligent technique for measuring the static or actual load of the axle, despite the presence of dynamic loads, source large number of errors, which affect the results of research currently carried out on dynamic weighing systems in operation, the reason why, until today, no system operating under normal traffic conditions, legally recognized, exists for the direct sanction of offenders.

The system designed around the aforementioned technique, object of the present invention, consists of the hard core of the complete high-speed dynamic weighing system by its coupling with the new information technology equipment existing on the commercial market such as: the camera system, vehicle license plate recognition software and data transmission by communication network. This global system allows the photos of the trucks to be taken automatically and the identification of data relating to the site of the measurement, the date and time of the measurement, the vehicle and its owner, at the expense of each axle and the overall vehicle weight. These data constitute the basic elements for the automated sanction of offenders in the event of overweight.

IV. Basic principles

The systems that fall within the technical field of the invention are designed to:

- Measure the static load of each axle of the vehicle;

- Measure the overall weight of the vehicle (PGV).

If we take an example of a rigid truck with 2 axles moving at normal speed on the traffic lane, the method of calculating the aforementioned parameters depends on the configuration of the sensors installed as follows:

- The sensors, which are installed transversely to the axis of the roadway, occupy the entire vehicle traffic lane; this configuration n ° 1 makes it possible to measure the load C exerted by each truck axle on the roadway and that the PGV is the sum of the measured loads of the two axles of the truck.

PGV = C (axle I) + C (axle!)

- The sensors, which are installed transversely with respect to the axis of the roadway, occupy the lane for the circulation of the wheels on the same side of the axle; this configuration No. 2 makes it possible to measure the load R exerted by each truck wheel on the roadway; the load of an axle is the sum of the measured loads of each wheel R of the same axle and that the PGV is the sum of the loads exerted by all the single or twin wheels of the truck which is also the sum of the measured loads of the two axles of the truck.

PGV = R (wheel 1) + R (wheel 2) + R (wheel 3) + R (wheel 4)

= C (axle 1) + C (axle2)

Taking into account the calculation method, the content of this invention is constructed on the basis of the load of any axle / vehicle by opting for the first aforementioned configuration. The same reasoning of one axle is valid for all the axles of the vehicle and that the PGV is a deduction from the measured loads of the axles of the vehicle.

V. State of the art

Overweight or administratively speaking "Technical overload" of vehicles is a major factor in the degradation of the road network leading to both a drop in the quality of the service offered to road users and a significant increase in the budgets allocated to road maintenance. pavements.

Also, and in addition to its detrimental effect on road safety, insofar as it contributes to putting the vehicle in a situation of instability and considerably reduces the capacities and efficiency of its safety devices, overweight also causes distortion. importance of fair competition between transport companies.

To fight against this offense, the Moroccan legislator has regulated the provisions governing the obligations and penalties relating to transport by road vehicles. Indeed, Law No. 52-05 on the Highway Code provides for provisions relating to the management of offenses related to exceeding the authorized Total Loaded Weight (GVW) and axle loads of transport vehicles, in particular:

- The obligation to measure the exceeding of the GVW and the axle loads by the use of the appropriate equipment;

- The introduction of proportional penalties according to the degree of exceeding the measured weight;

- The establishment of the co-responsibility of the owner of the vehicle, the shipper, the broker, the loader, the consignee or the principal having caused or participated in committing the offense of exceeding the authorized PTC by more than 40% or given orders to that effect.

To deter offenders, the overweight control of trucks is carried out by the police, using mobile axle scales as well as fixed dynamic weighing stations. If the use of mobile units can be done anywhere, and easy to deploy, fixed stations, for their part, have a more or less complicated operating principle since the weight measurement is carried out in two stages. The first is done through the running weighing system at high speed which constitutes the preselection system for potentially overweight vehicles, generally composed of two piezoelectric sensors, which are buried in the road, in one or two lanes perpendicular to the direction of the flow of road traffic. It makes it possible to measure the loads applied by the wheels on the roadway and then on the piezoelectric sensors. In fact, the vehicle applies longitudinal, horizontal and vertical forces on the road, through the wheels. The only component that is measured is the vertical force. The output of the sensor is an electrical charge which is proportional to the applied force.

The second is the precise control of the weight and on the basis of which the offender is fined. It is carried out through a fixed scale standardized and approved by the services responsible for legal metrology. This scale is embedded in a platform fitted downstream of the system in operation at high speed. Weighing operations are carried out statically or at low speed at less than 15 km / h.

According to the experience feedback, the two aforementioned measurement systems record a very low return compared to the resources mobilized for this purpose. In order to address this dysfunction, direct and automated repression is needed. The high-speed in-motion weighing technology described above fails to achieve the accuracy required for legal applications. Indeed, these systems measure instantaneous impact forces which are not equal to the static loads of the wheels and axles, but are affected by the dynamic interactions between the roadway and the vehicle, and can deviate from them by +/- About 30%.

In addition, and with a view to improving current systems, the research investigated is based on piezoelectric multi-sensor grids which measure vehicle axle loads under normal road traffic conditions. However, they do not make it possible to guarantee the metrological requirements required for the direct repression of overweight, although they consist of several sensors thus making a measurement distribution tending to have sufficient information on the dynamic load signal. The objective of this distribution of values is also to minimize, by calculating its arithmetic mean, the errors linked to the measurement of the static load of the axle. Therefore, the overall vehicle weight and the static axle load are estimated. In this regard, the calculation algorithms are based on statistical models which unfortunately require a distribution of measurement values of very large size in order to estimate an approximate value of the static load of the axle, the precision of which is strongly linked to the '' standard deviation of the values constituting the distribution depending on:

- The uni or profile of the roadway;

- The speed of the vehicle;

- The vehicle's suspension system;

- The quality of the piezoelectric sensor.

In relation to the technology of piezoelectric sensors, by piezoelectricity is meant the physical property that certain materials have to electrically polarize when they are subjected to a mechanical stress (direct effect) and to deform mechanically when they are subjected to a electric field (indirect effect). The piezoelectric material is characterized by its piezoelectric constant which reflects the proportionality between the deformation and the electric displacement at zero or constant electric field. Quartz remains the piezoelectric element which has, in addition to other mechanical and electrical qualities, better precision. Due to the weak electrical signals that they produce, following a variation of constraints, the piezoelectric sensors are not used alone but they are coupled with acquisition chains, the main element of which is the charge amplifier. Coupling with this electrical component has been strongly recommended in the literature due to the capacitive nature of piezoelectric sensors.

Finally, find a method for reconstituting the load on the vehicle axle, define the optimal number of piezoelectric sensors and their inter-distances constituting the sensor grid, process the reconstituted signal of the axle load and compensating for the dynamic loads generated by the movement of the vehicle are the main axes of the present invention detailed in this document.

In view of the above, it should be noted that at present, no dynamic weighing system operating at high speed operating under normal traffic conditions exists on the market for the direct sanction of offenders, and this had with regard to significant errors in the measurement of the static load of the axle and consequently of the overall weight of the vehicle.

VI. Description of the invention

The present invention relates to the design and production of a high speed dynamic in-motion weighing system for measuring the static axle load / and the overall vehicle weight (PGV) under normal traffic conditions. road. This invention is not based on the statistical computational approaches to measure the static axle load and the PGV but rather on the basis of the analog signal processing of the axle / vehicle, reconstructed using a precise algorithm presented later. .

The technical solution that is the subject of the invention is based on the following components:

1. The algorithm for reconstructing the analog signal of the axle / vehicle;

2. The design and definition of the optimal grid of piezoelectric sensors;

3. The design and production of an analog filter comprising the following electronic cards:

- Signal conditioning card produced by the piezoelectric sensor;

- Axle load analog signal processing card and static load measurement.

Figure 1 gives an overview of the system object of the present invention. He understands :

- The piezoelectric sensor grid (1);

- The control unit (2); within this block sits the various electronic cards, the microprocessor and the acquisition card.

The additional equipment that is required to have a valid system for automated control and sanctioning is the new information technology tools that exist on the commercial market, in particular:

- The digital photo-taking camera;

- Plate recognition software;

The communication network, data security;

Computer accessories such as servers, routers ... at. Axle / vehicle analog signal reconstruction algorithm

Let n + 1 pairs (to, vo), (ti, vi),. , (t _n , v _n ) (3) where tk is the passage time of the axle / vehicle (5) on the sensor of order k, installed in the roadway on the traffic lane (6), and the V _k is the voltage generated by the sensor of order k following the instantaneous force exerted by the axle / on this sensor. These data are taken against a defined reference (time, voltage) (4) as shown in Figure 1.

The analog signal (7) of the axle load / product, following the passage of the vehicle (5) over the sensor grid (1) (Figure 1), follows the shape of the Lagrange polynomial interpolated from the n + 1 above-mentioned couples.

In order to reduce interpolation errors, the positions of the sensors are well defined in the sensor grid (1) according to the Chebyshev formula defined below.

Equation (1.1) represents the mathematical formula of the Lagrange polynomial expressed in the time domain.

F is the Lagrange polynomial of at most degree n which passes over the points (t _k , V _k ) (3), with t is the time of passage of the axle / at any point M (f, v) of the sensor grid (1) along the longitudinal axis of the roadway (along the time axis) (8) and vest the interpolated voltage at this point such that F (t) = v.

The value of F (t) multiplied by the calibration coefficient defined in equation (1.9) represents the instantaneous force exerted by the axle / at point M. The Lagrange polynomial fulfills the following conditions:

It is very difficult to determine the degree of a polynomial that should be stopped in the interpolation of a function. A higher degree polynomial is not always the best choice because it can produce unstable results.

The instability mainly stems from:

The choice of interpolation points; equidistant or uniform points do not necessarily give better results; The function whose variations are slow, already smoothed, is better interpolated by a polynomial of high degree than by a polynomial of low degree, while the function whose derivatives change rapidly is better interpolated by a polynomial of low degree whose practice demonstrates a degree less than or equal to 3;

Algorithm used whose qualities depend on the mathematical method from which derives.

Equation (1.3) represents Chebyshev's formula expressed in the spatial domain specifying the positions of the piezoelectric sensors:

X _k is the position of sensor k in the grid (1) which contains n + 1 sensors. The numbers a and b are the abscissas of points A and B, on the longitudinal axis of the road traffic lane (6) where the sensors are installed, constituting the terminals of the sensor grid (Figure 1). Therefore, the grid is defined by two points A (a, 0) and B (b, 0) having a length of (ba).

To obtain the best possible estimate of the function F, it is necessary to choose the n + 1 points of interpolation xo, xi ... x _n so as to minimize the maximum on [a, b] of the function

N. B: It should be noted that the passage from the time domain to the spatial domain on the longitudinal axis of the traffic lane or vice versa, it suffices to intervene the average speed of the vehicle that can be measured (known grid length and time of passage of an axle on the grid (1) can be measured by the timer of the data acquisition card used in the technical solution).

The main conclusion drawn, through the signal reconstruction method explained previously, is its proven efficiency and performance which is based on the Lagrange polynomial interpolation algorithm, coupled with Chebyshev conditions with a sampling step non-uniform according to equation (1.3).

This algorithm only takes into account the limited band of the signal in the time domain, being the interval of definition of the force F (t), and has no special requirement for the estimation of the signal frequency spectrum or infinite distribution of elements used to use the signal reconstruction method.

The dimension of the formed torque distribution (t _k , V _k ) (3) for which the reconstruction of the axle load signal / gives better results is from n = 10. Hence the minimum number of sensors constituting the grid is 10 sensors. However, for reasons of economic profitability of the system, the optimum number of sensors is 10 sensors. b. Sensor grid (1)

The spectral analysis of the axle load signal / shows the existence of two fundamental frequencies fi and respectively located between 1-3 Hz and 6-15 Hz. These frequencies correspond to the movements of the suspended and unsprung masses of the vehicle. (5). The values of these frequencies do not take into account the quality of the uni, but they depend on the speed and the quality of the vehicle's suspension system (5). Their amplitudes are linked to both the speed of traffic and the quality of the road surface (6).

The design of the sensor grid (1) must be designed in such a way that the axle load signal / is reconstructed, from the values generated by the sensors, on a limited band and this in order to have better accuracy of the signal reconstruction by Lagrange polynomial.

Furthermore, the grid of piezoelectric sensors (1), installed on the roadway (6), forming part of the dynamic weighing system in motion at high speed, object of this invention, does not make it possible to produce a continuous time signal but rather a continuous time signal. distribution of measurements taken at a time series determined at a non-uniform sampling interval according to equation (1.3) which is the basis for the design of the sensor grid (1). Figure 1 illustrates the design of the sensor grid (1). vs. Dynamic load compensation

Once the analog signal of the axle / vehicle load (5) is reconstructed, the next action taken is to process the reconstructed signal through an electronic device. This device is designed and produced in order to eliminate or compensate for the variable component of the signal corresponding to the dynamic loads. generated by the axle / and then measure the DC component which is the static axle load / ^' of the vehicle.

Indeed, the reconstructed analog signal of the axle / vehicle load (5) is generally an aperiodic signal. Equation (1.1) can be reformulated as follows:

With :

- Q is the instantaneous force applied by the axle / of the vehicle (5) on the road surface (6) (Axle load signal / ^' );

- P is the static axle load / (DC component of the signal);

- K is the stiffness of the vehicle's suspension system (5); An the oscillation amplitude of the axle load of the ^nth frequency of the signal; f _h is the phase at t = 0 (initial conditions of oscillation) of the n ^th frequency of the signal;

- sin (co _n t + (p _n ) is the variable component of the analog signal which

0 represents the dynamic loads of the vehicle axle (5) generated in particular because of:

• Uneven pavement (poor level of pavement) (6);

• Vehicle suspension system (5);

• Vehicle speed (5).

The reduction of errors due to dynamic loads in dynamic weighing systems in operation at high speed requires prospecting, instead of investigating in statistical approaches currently being researched, the results of which cannot be controlled which strongly depend on variations in loads. dynamics (variable component of the analog signal) other techniques. A new intelligent technique has been invented which tends to isolate or compensate for the variable component of the reconstructed analog signal of the axle / vehicle (5) and then to measure the DC component which corresponds to the static load of the vehicle. vehicle (5) which it is desired to measure with better precision. In this regard, the technique introduced is based on the “Zero Crossing Detector” device, considered to be the hard core of the analog filter produced within the framework of the present invention.

The analog filter is based on performing the operations relating to:

- Conditioning of the signal produced by the piezoelectric sensor in view of its low power. This conditioning allows the exploitability of the values produced by the sensor which are in the form of voltage and consequently it will be possible to read a maximum value (deterministic value) given by the sensor, difficult to acquire since the signal is often too noisy whose signal to noise ratio depends on the quality and nature of the sensor. The maximum value corresponds to the maximum instantaneous force applied by the axle / on the road surface (6);

- Reconstructed analog signal conditioning of the axle / vehicle load (5) to adjust the signal amplitude to ensure the efficiency of the analog filter operation;

- Elimination of the DC component of the reconstructed signal of the axle load / vehicle (5) by introducing a filter ¹ high-pass type of order;

- Process the output signal of the ^1st order filter using the “Zero Crossing detector” technique.

Definition of "Zero Crossing detector": Zero Crossing is defined as the point where a mathematical function changes sign from positive to negative or vice versa. In other words, they are the intersection points of the curve of a continuous function with the x-axis.

The application of the detector of the roots of the mathematical function on the filtered analog signal of the axle / of the vehicle (5) is likely to reveal the temporal measurements which coincide with the zero values of the dynamic loads. By neglecting the errors due to the electronic components compared to those of the dynamic loads, the values measured on the reconstructed analog signal of the axle load / ^' , with the time measurements thus identified by this device, correspond well to the voltages produced by the static axle / vehicle loads thanks to dynamic load compensation. To calculate the static axle load / ^' , it suffices to multiply the measured tension by the calibration coefficient defined previously using equation (1.9). d. Electronic cards designed and produced i. Signal conditioning card produced by the piezoelectric sensor

Each piezoelectric sensor of the gate (1) has its own electronic circuit designed for conditioning the signal produced by this sensor. This electronic circuit is composed of two stages in cascade made up of:

The ^1st stage, illustrated in Figure 2 (a), consists of a charge amplifier considered as a means of converting the charge generated by the piezoelectric sensor into a signal in the form of an exploitable voltage. A charge amplifier uses the basic integrator topology. For this, the operational amplifier Ai is mounted in the circuit with a feedback consisting of a capacitor C1 mounted in parallel with a resistor R3 preventing the amplifier from reaching its saturation level. The output voltage of the ^1st stage is proportional to the integral of the input current according to equation (1.6).

When the charge Q produced by the piezoelectric sensor varies as a function of time, the voltage signal v _a at the output of the ^1st stage of the conditioning circuit comprises a DC component and a dynamic component. Equation (1.6) is therefore in the form of:

Where Vo is the DC component of the output signal of the charge amplifier which is considered to be the target value to be measured in the 2 ^nd stage of the conditioning circuit. The 2 ^nd component of equation (1.7) corresponds to the noise of the signal.

The operational amplifier Ai is supplied by direct voltages of + 11V and -11V, the supply current of which is limited via resistors R1 and R2 installed between the power source and the amplifier.

As the charge produced by the piezoelectric sensor is too low, we can consider that v ₀ (t) is one of the weak signals.

The 2 ^nd stage, shown in Figure 2 (b), is constructed using an emitter-follower circuit based on a transistor T (npn type) which operates in the small signal regime. The output of the emitter of transistor T is connected to the capacitor of C2 decoupling in parallel with resistor R4. The decoupling capacitor C2 makes it possible to separate the point of polarization of the transistor and the small existing signals in dynamic mode. Its impedance is negligible for small signals and infinite for bias currents.

Indeed, suppose that C2 is disconnected (open circuit), the current of the emitter i _e varies as a function of the collector current i _c . This variation of i _e causes a variation in the voltage across resistor R4, which tends to decrease the polarization of junction VBE at the same rate as the variation of i _e .

In the case where C2 is connected, the capacitor eliminates the dynamic component, then then the voltage between the terminals of the resistor R4. As a result, the voltage Vs is stable, which corresponds to the voltage Vo reduced by 0.7 V corresponding to the base-emitter junction voltage VBE according to equation (1.8).

V = V ₀ -V _BE (1.8)

Hence the output voltage Vs of the electronic signal conditioning circuit depends on the output voltage Vo of the charge amplifier which is a function of the charge Q produced by the piezoelectric sensor.

Although the signal at the input (9) of the conditioning card is too noisy, it makes it possible to acquire at the output of the ^2nd stage a deterministic signal whose maximum value varies proportionally with the instantaneous force applied by the wheel on the piezoelectric sensor. ii. Axle load analog signal processing card / and static load measurement

This electronic board contains the following electronic components as shown in Figure 3:

The axle load analog signal processing board / and the static load measurement is considered to be the mother board of the dynamic weighing system in operation at high speed since at its level all calculations are carried out. of the overall system.

After the axle / vehicle (5) has passed over the piezoelectric sensor grid (1), the analog signal processing board of the axle load / and the static load measurement reads data, via a microprocessor (10), generated by the piezoelectric sensors in the form (ί _/ , n,) relating to the passage times of the axle / vehicle on the piezoelectric sensor grid (1) as well as the corresponding voltages generated. Once the data is read by the microprocessor (10), having a high computing power, executes the algorithm stored there, relating to the reconstruction of the analog signal of the axle / vehicle, developed on the basis of the 'interpolation by Lagrange polynomial. The signal generated by the microprocessor (10) on its DAC (Digital analog converter) port is the reconstructed analog signal of the axle load / applied to the roadway (6).

After reconstructing the analog signal of the axle / vehicle (5), this signal passes through a circuit formed of three stages in cascade, namely:

A. ^1st stage: Amplification of the reconstructed analog signal

To improve the exploitability of the reconstructed signal, the signal transmitted by the microprocessor (10) via its DAC port is amplified thanks to the amplification stage illustrated in Figure 3 (c) using for this purpose an operational amplifier A2 (mode multiplier). This stage makes it possible to obtain a physically usable analog signal (V2) and, consequently, to reduce measurement errors. The amplitude of the analog signal can be varied using the potentiometer (POT).

B ^2nd stage: Filtering of the output signal of the amplification stage

The output signal of the amplification stage is filtered in order to extract the DC component relating to the static load of the axle / and to pass the variable component corresponding to the dynamic loads. This operation is implemented by the 1 ^st degree "high pass" type filter, shown in Figure 3 (d), made up of capacitor C3 and resistor R9 which is connected to ground. The cutoff frequency of this filter is 0.16 Hz.

C. ^3rd floor: Compensation for dynamic loads

The signal delivered by the filtering stage is an aperiodic signal whose average is zero (abscissa axis), we proceed to the last operation based on the “Zero Crossing detector” technique, the circuit of which is shown in Figure 3 ( e), which consists in producing a positive square signal. To do this, an operational amplifier A3 of the comparator type is used, the output of which is connected to the power source of this amplifier by the resistor R10. The operational amplifier is supplied with a single power source of + 5V. Thus, the aperiodic signal produced by the filter is rendered by this stage a square signal whose projection points of the sides (rising and falling) with the abscissa axis coincide with the points of intersection of the aperiodic signal (filtered signal) with this axis.

Finally, to enable the data on the amplified aperiodic signal to be recorded, from the output of the ^{aforementioned 1st} stage, which correspond to the projection points of the sides of the square signal on the abscissa axis, an Arduino type acquisition card is integrated in the circuit of Figure 3 which executes the program recorded therein.

In order to monitor the transmission of energy in the electronic circuit of the axle load analog signal processing card / and the static load measurement, this card is equipped with ten LEDs from D1 to D11 which mark the state "0" or "1" according to the following modes:

Leds D1 to D10 light up sequentially and then go out, which proves that the microprocessor (10) has successfully read all the data transmitted by the signal conditioning card produced by the piezoelectric sensor;

D1 and D5 turn on and then turn off, which means that the execution of the Lagrange polynomial algorithm is successfully completed;

D6 and D10 light up then go out, which proves that the production of the analog signal via the DAC port of the microprocessor (10) has been carried out successfully; D11 lights up then goes out, which shows that the acquisition card has calculated the static load of the axle / ^' of the vehicle (5). e. Method of calculating the static load of the vehicle axle i

A linear relationship between the voltage coming from the acquisition card included in the "Axle load analog signal processing card / and static load measurement" and the actual axle load is the basis of static load calculations. This linearity requires that all stages of the signal production and processing process have a proportional relationship between the input signal and the output signal, in particular the conditioning, amplification, filtering and “Zero Crossing” circuits. detector ”. However, this condition cannot be realized in practice due to the systematic errors of each component constituting the overall process, which ultimately leads to producing a cumulative error at the output of the process. The value obtained by the acquisition card is in the form of voltage which is multiplied by the calibration coefficient in equation (1.9) below making it possible to transform these values into a unit of weight.

The calibration coefficient expresses the ratio between the true value of the axle load / expressed in kilograms and the measured reference value of this axle expressed in volts according to equation (1.9). c True value (weight)

^ calibration (1.9)

Reference measured value (voltage) f. Brief description of the drawings

Figure 1: The described diagram of the system object of the agreement which consists of:

The piezoelectric sensor grid (1);

The control unit (2) which houses the electronics of the dynamic weighing system in operation at high speed which is the subject of the invention.

Figure 2: The diagram describes the electronic circuit of the "Signal conditioning card produced by the sensor" which is made up of two stages:

(a) Conditioning of the signal produced by the piezoelectric sensor (a);

(b) Generation of a deterministic signal (b).

Figure 3: The diagram describes the electronic circuit of the "Axle load analog signal processing board / and static load measurement" which is made up of three stages:

(c) Amplification stage (c);

(d) Filtering stage (d);

(e) Dynamic load compensation stage (e).

Claims

1) Dynamic weighing system in operation at high speed, is a device which measures the static load of the axles and the overall weight of the vehicle (5) under normal road traffic conditions. This system is made up of: i. A grid of piezoelectric multi-sensors (1) whose minimum number of sensors is set at 10 sensors arranged transversely to the axis of the road at a non-uniform pitch; ii. An analog filter that processes the analog signal from the vehicle axle (5) to compensate for dynamic loads and then measure the static axle load and overall vehicle weight (5).

2) A dynamic running weighing system at high speed, according to claim 1, characterized in that the operation of measuring the static load of the axle and the overall weight of the vehicle (5) is based on the reconstruction of the signal analog of the axle load of the vehicle (5) over a limited band. The shape of the reconstructed analog signal follows the form of the following equation (1.1):

With F is the Lagrange polynomial of at most degree n which passes over the points (t _k , V _k ) (3). t _k is the time of passage of the vehicle axle of the vehicle (5) on the sensor of order k and V is the voltage generated by the sensor of order k following the instantaneous force exerted by the axle of the vehicle on this sensor.

3) Dynamic weighing system in motion at high speed, according to claim 1, characterized in that the grid of piezoelectric sensors (1) contains at least ten piezoelectric sensors, arranged transversely with respect to the axis of the roadway of the vehicle ( 6), at a non-uniform pitch. The position of each sensor, along the abscissa axis, is defined by the following formula:

With :

- (n + 1) is the number of piezoelectric sensors

- a and b are the abscissas of two points A and B delimiting the grid of piezoelectric sensors (1) along the abscissa axis (parallel to the longitudinal axis of the roadway (8).

4) A high-speed running dynamic weighing system according to claim 4, characterized in that the length of the piezoelectric sensor grid (1) which contains at least ten piezoelectric sensors is at least 1.90 meters. This length makes it possible to record at least one zero crossing or cancellation of the variable component of the analog signal of the vehicle axle, and this after eliminating the DC component of said signal;

5) Dynamic weighing system in operation at high speed, according to claim 1, characterized in that the analog filter consists of a "card for conditioning the signal produced by each sensor" and a "card for processing the signal analogue of axle load and static load measurement ”;

6) Dynamic weighing system in operation at high speed, according to claim 6, characterized in that the "card for conditioning the signal produced by the sensor" is formed of a two-stage electronic circuit: the ^1st stage is formed a charge amplifier, consisting of the operational amplifier Ai with a feedback formed by the capacitor C3 mounted in parallel with the resistor R3. This stage is coupled with the 2 ^nd stage which is formed using an emitter-follower circuit based on a transistor T (npn type) whose voltage at the emitter follows the input voltage. The output of the emitter of the transistor is connected to the decoupling capacitor C2 in parallel with the resistor R4. Even with a signal that is too noisy at the input, this card makes it possible to acquire at the output of the 2 ^nd stage a deterministic signal whose maximum value varies proportionally with the instantaneous force applied by the wheel on the piezoelectric sensor; 7) Dynamic weighing system in motion at high speed, according to claim 6, characterized in that the "Card for processing the analog signal of the axle load and the measurement of the static load" consists of a electronic circuit formed of three stages mounted in cascade: "Amplification stage", "Signal filtering stage", "Dynamic load compensation stage via Zero Crossing detector";

8) Dynamic weighing system in operation at high speed, according to claim 8, characterized in that the amplification stage consists of: an operational amplifier A2 of the multiplier type, four resistor R5, R6, R7 and R8 and a potentiometer (POT). This stage allows the amplitude of the analog signal of the rebuilt vehicle axle to be varied in order to increase its usability and reduce measurement errors;

9) Dynamic weighing system in operation at high speed, according to claim 8, characterized in that the signal filtering stage: The "high pass filter" is formed of an RC circuit (R3, C9) having the frequency cutoff less than 0.2 Hz. This filter removes the DC component of the analog signal from the vehicle axle (5). Under the effect of the filter, the analog signal oscillates, aperiodically, around the x-axis. This result allows the 3 ^rd stage of the “Axle load and static load measurement analog signal processing card” to play its main function of canceling dynamic loads;

10) Dynamic weighing system in operation at high speed, according to claim 8, characterized in that the dynamic load compensation stage via "Zero Crossing detector" is formed around an operational amplifier type A3 comparator and a resistor R10 placed between the output of amplifier A3 and its supply voltage. The operational amplifier is powered from a single + 5V power source. The integration of this stage makes it possible to identify the temporal measurements which coincide with the zero values of the dynamic loads.