Just get rid of the bridge etc. and use a diode, zener and 1K resistor from the gate to MT2 of the triac, with the 47K gate to MT1.
Use two of those simpler circuits connected opposite polarities. You can use thyristors instead as each device is only handling a single polarity.
Or, use an opto-triac as the trigger after a bridge, with it's triac output connected between MT2 and gate through a low-value resistor, eg. 10 ohms.
That keeps it to a single triac.
A generic opto-triac trigger circuit:
https://i.stack.imgur.com/c3Jkp.png
Make sure it is an instantaneous type, there are also zero-crossing versions for low interference power control, that only turn on at the start of a half cycle.
Though I must reiterate that the application to try and control the output of a "dynamo" (alternator) seems very flawed in practice, as dknguyen says - a simulation as you have will not allow you to see the real-world effects and problems.
"Crowbar" circuits like this are normally last-resort-only and never as any kind of regulation. Filtering and series / switched-mode regulation is vastly more practical in real-world designs.
And note that if you are using some form of "active" rectifier for efficiency, with some forms shorting the input mid-cycle can cause damage or high current surges as the output is shorted back though the active switches.
Clipping transients, with either a suppressor or transistor + zener based circuit, is also much better as it only absorbs the _difference_ between the normal and transient level outputs, not the entire source power; the output to whatever load is not shut down.